Student Projects Evidence Based Clinical Practice Guideline

The reliability and validity of fundal height measurement

Anne Charles

Anne Elizabeth Charles, CNM


Abstract
One of the principal aims of prenatal care is the detection of intrauterine growth abnormalities and the pregnancy complications that they might represent, such as fetal growth disturbance, multiple gestation, amniotic fluid disorder and erroneous dating. One method of clinical assessment of fetal growth is the fundal height measurement; a widely used technique involving measurement of the maternal abdomen from the symphysis pubis to the uterine fundus with a tape measure during the second and third trimesters of pregnancy. While the measurement is most often used to determine fetal growth, other variables including amniotic fluid volume, maternal adipose, fetal lie and presentation, can influence the measurement obtained. Fundal height measurement is considered valuable because it may improve detection of fetal growth abnormalities and thus allow monitoring and treatment to reduce perinatal morbidity and mortality. It is acceptable to women because it often provides reassurance of fetal growth and well-being, is easily obtainable, inexpensive and relatively non-invasive. It has become an integral aspect of midwifery care. However, it is well documented that, despite considerable and careful examination of fundal height, fetal growth abnormalities are not always recognized antenatally. Research has raised some questions about the technique and usefulness of fundal height measurement. The predictive value has been studied by several investigators who found limited evidence to support its use ( Lindhard, 1990 ) and others who found problems related to interexaminer reliability and clinician bias.( Engstrom, 1994 ) It is the aim of this study to consider the literature and to determine the validity and reliability of this commonly performed intervention.

Clinicians are increasingly expected to adopt evidence-based practice, using research findings to make informed decisions and influence their actions.( Rooks, 1999 ) Evidence based practice can help to ensure that interventions are clinically appropriate, cost effective, and result in positive outcomes for clients. In order to improve the reliability and validity of fundal height measurement, an evidence based clinical practice guideline has been produced following evaluation of available research. By promoting continuity and uniformity of fundal height measurement technique, fetal growth abnormalities may be more accurately and promptly detected, appropriate interventions initiated, and neonatal outcomes improved.


Text
Introduction to the problem

One of the principal aims of prenatal care is the detection of intrauterine growth abnormalities and the pregnancy complications that they might represent, such as fetal growth disturbance, multiple gestation, amniotic fluid disorder and erroneous dating ( McGeown, 2001).Current methods of clinical assessment for fetal growth abnormalities include manual abdominal palpation, ultrasound scanning and measurement of the fundal height. The fundal height measurement is a widely used technique involving measurement of the maternal abdomen from the symphysis pubis to the uterine fundus with a tape measure during the second and third trimesters of pregnancy. It is easily obtainable, inexpensive and relatively non-invasive but the technique has been questioned as to its usefulness and reliability. Obtaining and recording this measurement takes considerable time during each prenatal visit and therefore it is important to examine the usefulness of this technique. It is the aim of this study to review the current literature, to consider the techniques used and to determine the validity and reliability of this intervention commonly performed by midwives. In order to justify the universal use of the fundal height measurement, midwives must ensure that the technique is worthwhile, achieves its purpose, and can be dependably reproduced.

Scope of the problem.

The American College of Obstetricians and Gynecologists (A.C.O.G.) and the Public Health Service Expert Panel on the content of prenatal care advocate the use of fundal height measurement and recommend that it be obtained at routine prenatal visits ( Public Health Service, 1989 ). However, there are no published standards in the United States that describe the technique. It was thought that the widespread introduction of fundal height measurement would improve the detection rate of fetal growth abnormalities but the results have varied widely from studies undertaken at different centers ( McGeown, 2001 ).

The ability to accurately predict pregnancy complication using fundal height measurement may be affected by a number of factors including the technique used, clinician skill, the number of clinicians involved, the maternal position, the status of the maternal bladder and the possibility of clinician bias. There are studies which support the value of fundal height measurement ( Mathews, 1997 ) and some which have suggested improvements to ensure uniformity, ease of interpretation and an improvement in predictive value ( Gardosi, 1999 ). The scope of this issue is vast when one considers that three to four million women become pregnant each year in the United States and that the majority of them will undergo fundal height measurement at each prenatal visit during the second and third trimesters of pregnancy.

Significance to midwifery and women's health.

Fetal growth abnormalities and the pregnancy complications that they may represent have a significant impact on women's health and the outcomes of their pregnancies. Clinicians have not always been able to detect growth problems prenatally resulting in the potential for perinatal morbidity and mortality. Early diagnosis allows care to be planned accordingly in order to reduce perinatal compromise. If an inexpensive, relatively non-invasive, effective intervention can improve the detection rate then it should be provided ensuring reliability, continuity and efficaciousness.

The diagnosis of a fetal growth abnormality and the surveillance and management prescribed may create additional stress and, perhaps, a financial burden to the woman and her family. Activity restriction, dietary modification and even hospitalization may be deemed necessary. If undiagnosed, and 65% of cases of intrauterine growth restriction ( IUGR ) are not identified prenatally ( Mc Farlin, 1994 ), the impact on the woman and her family may be significant if long term health problems ensue.

The American College of Nurse Midwives' philosophy of midwifery states that "..the practice of nurse midwifery encourages continuity of care; emphasizes safe, competent clinical management..." and therefore midwives must ensure that any component of the care they administer is both valid and reliable. That care must also be research based and effective.

A fundal height measurement is obtained at each prenatal visit during the second and third trimesters of pregnancy and this intervention may take a significant proportion of the average 15 - 22 minute time period that midwives typically have to assess prenatal clients. It is a technique that must be learned to ensure reliability and continuity and therefore both the learning and the teaching are significant to midwifery. If a fundal height measurement deviates from the norm the midwife may become involved in sanctioning additional interventions such as ultrasound. In some clinical settings, a diagnosis of IUGR or macrosomia may even necessitate transfer of a woman's care to an obstetrician or perinatologist. It is the aim of this study to consider the literature and to determine the validity and reliability of this commonly performed intervention.

The gap between current and optimal evidence based practice.

There is undoubtedly a significant gulf between current clinical practice and optimal evidence based practice when one considers fundal height measurement and its role in detection of fetal growth abnormalities. Evidence based practice indicates that the care is based on findings from clinical studies that were designed to measure the effectiveness of specific therapeutic methods, especially systematic reviews that consider the strength and quality of evidence supporting or refuting a particular practice and lead to a consensus regarding guidelines ( Rook, J 1999 ).

While it is an appealing intervention given the fact that it is readily available, easy to perform and relatively non-invasive, the reliability and validity of fundal height measurement have been questioned. ( Engstrom, 1993 ; Lindhard, 1990; Neilson, 2001 ) In order for fundal height measurement to be both reliable and valid, there must be continuity and uniformity of technique. Research suggests that outcomes can be affected by several modifiable factors including the number of clinicians involved, clinician bias, maternal position, and status of the maternal bladder. Health outcomes may therefore be adversely affected if there is a lack of continuity and uniformity of technique. Fetal growth abnormalities such as SGA or LGA may be overlooked or perhaps even over diagnosed leading to increased maternal anxiety and the possibility of unnecessary interventions.

Fundal height measurement in itself is an inexpensive intervention which contributes to its widespread appeal and use. However, if all women undergo the procedure at each prenatal visit during the second and third trimesters, the cost is significant in terms of clinician time and therefore, to justify its use, it must be shown to be reliable and valid. If the measurement is not, however, additional costs may be incurred by the woman and her family including those related to unnecessary medical interventions and those resulting from neonatal care necessary for undiagnosed IUGR or macrosomia.

Provider and client satisfaction has been addressed through research ( Engstrom 1993 ) and it is interesting to note that when considering maternal position for fundal height measurement, clinicians preferred the supine position because of ease of measurement and women were more comfortable in a position with trunk elevation and knee flexion. There can be some conflict between client and clinician preference.

Theoretical and operational definition of terms.

Theoretical definition.

Theory is the ultimate aim of science: it transcends the specifics of a particular time, place and group of people and aims to identify regularities in the relationships among variables ( Polit, Beck & Hungler, 2001 ). Fundal height measurement is a commonly used clinical intervention to screen for fetal growth abnormalities during the second and third trimesters of pregnancy. Fundal height uses landmarks on the maternal abdomen as points for a measurement that theoreticaly reflects fetal growth and well-being. Because of the anatomic and physiologic changes that take place in the uterus during pregnancy, the fundus becomes evident abdominally and therefore measurable. As it is not feasible to measure the fetus directly at prenatal visits, theoretically the growing fundus becomes a marker variable for fetal growth.


Operational definition.

An operational definition clarifies and expounds how the variables of a quantitative study will be observed and measured in the research situation. Fundal height should be measured with a non-elastic tape measure from the top border of the symphysis pubis to the uterine fundus in the midline, with tape curving around the fundus to its apex. The maternal bladder should be empty and her position should be supine with her legs extended.


Reliability is defined as the repeatability, reproducibility, or consistency of a measurement, when the measurements are obtained under identical conditions. ( Engstrom, 1993 ) In order for a fundal height measurement technique to be considered reliable it must be carried out in a consistent and uniform manner. The reliability of various fundal height measurement techniques has not been studied extensively yet reliability is essential if the measurement is to prove useful in clinical practice. Both intra and inter-examiner reliability may be determined by comparing the measurements obtained by different clinicians on a variety of subjects under the same conditions.


Validity is the degree to which an instrument measures what it is supposed to be measuring. ( Polit, Beck & Hungler 2001 ) Validity of a fundal height measurement technique may be determined by a criterion-related validity assessment when the researcher seeks to establish the relationship between scores on an instrument and some external criterion.

Assumptions and philosophical approach.

In English, the word midwife means "with woman" and in French, she is called "sage femme" meaning "wise woman". The philosophy of the profession reflects these descriptions. The midwife has a unique role which is complementary to but different from that of other health care professionals in the care of women and babies. The European Community Midwives' Directives state that one of the principal activities of a midwife is " to monitor normal pregnancies; to carry out examinations necessary for the monitoring of the development of normal pregnancies;" ( Mayes, 1989 ).The philosophy of the American College of Nurse-Midwives further describes the role - " The practice of nurse-midwifery encourages continuity of care; emphasizes safe, competent clinical management; advocates nonintervention in normal processes; and promotes health education for women throughout the childbearing cycle." ( Varney, 1997 ).

Fundal height measurement fits both descriptions given that research suggests that, when carried out with continuity of technique, it is a non invasive way to serially monitor fetal growth, detect growth abnormalities and provide the woman with reassurance. The assumption that fundal height measurement correlates with fetal growth has been both supported ( Euan et al, 1995; Engstrom, 1993) and refuted ( Lindhard et al, 1990; Neilson, 2001 ).

By far the majority of women in the US give birth in a hospital setting. Over the past three decades, there have been great advances in technology in an attempt to improve pregnancy outcomes. Obstetric practice has also changed and medical intervention is now more commonplace ( Mayes, 1989 ). Increased medical control of childbirth has changed the role of the midwife in many respects resulting in a decreased opportunity to exercise the degree of clinical expertise and responsibility for which she/he trained. Fortunately, evidence based clinical practice has caused many interventions to be challenged and practices such as episiotomy and continuous electronic fetal monitoring have been found to be questionable when subjected to close scrutiny ( Varney, 1997 ). Fundal height measurement represents an easily obtainable, relatively non-invasive and inexpensive intervention that fits the midwifery philosophy. There remains, however, a consensus among researchers that further studies are warranted to support the assumption that it can reliably reflect fetal growth.

Limitations of the state of the science in this area.

Numerous studies have addressed fundal height measurement and its role in prenatal care and detection of fetal growth abnormalities. Research remains in its infancy and the majority of authors acknowledge the need for further studies
(Engstrom, 1993; McGeown, 2001; Neilson, 2001.) The largest randomized controlled trial ( Lindhard, 1990 ) was unable to demonstrate that fundal height measurement increased detection of fetal growth abnormalities but the authors do acknowledge the possibility of clinician bias and the need for further research. The majority of the other studies considered were smaller and not randomized yet they are important and there is a consensus that research should be ongoing.

Chapter 2.

The fundal height measurement is a widely used tool to aid the detection of fetal growth abnormalities during the second and third trimesters of pregnancy. While it is an appealing intervention given the fact that it is inexpensive, relatively non invasive and easy to perform, its reliability and validity have been questioned.

Since the early part of the twentieth century measurement of the fundal height with a tape or pelvimetry caliper has been advocated ( Engstrom, 1993 ). Traditionally, fetal growth has been assessed by comparing the height of the uterine fundus to landmarks on the maternal abdomen. Indeed, this technique is still widely used by midwives in Great Britain. Observant clinicians as early as Sutugin in1875 ( Engstrom, 1993 ) noted significant variation in the distance between maternal abdominal landmarks and recommended more formal measurement. He found that the measurement between the maternal symphysis pubis to the umbilicus could vary from 15 cm to 22 cm. Other researchers reaching similar conclusions were McDonald in 1906 and Smibert in1962 ( Engstrom, 1993).

It is the aim of this study to consider more recent research, to explore various techniques suggested and to determine whether the fundal height measurement is a reliable and valid aspect of midwifery care.


Review of the literature.

Studies that support the use of fundal height measurement.

Mathews and Prabha ( 1987 ) compared three methods of measurement of the maternal abdomen during pregnancy to determine which was the most reliable in screening for SGA. Fundal height, abdominal girth and abdominal length were measured during routine prenatal examinations to assess the reliability of each method in screening for low birth weight for gestation infants who accounted for 72% of perinatal losses in one unit in South India between 1977 and 1982.

Standard graphs were constructed based on 584 sets of observations from 250 women with the following criteria between September 1983 and April 1985.

1) singleton pregnancy with known gestational age.( The authors do not state how the gestational age was determined yet they acknowledge that one third of their clients did not know their LMP.)
2) absence of medical or obstetric complication.
3) delivery at term in the Christian Memorial Hospital, Vellore, India.
The graphs based on the data were used prospectively to screen for low birth weight for gestation in high and low risk pregnancies.

Mean fundal height , and they do not clearly state how this was calculated , abdominal girth measured at the level of the umbilicus and the mean ratio of fundal height to abdominal length were calculated at two week intervals between 20 and 27 weeks and at weekly intervals between 28 and 40 weeks gestation. A non elastic tape was used and all 3 measurements were obtained by medical staff with varied clinical experience, which could potentially influence the outcomes, but all were instructed on a standard technique. The mean and 1 and 2 standard deviations above and below the mean for each variable were plotted against the appropriate geatational age and smoothed free hand curves were drawn to produce the graphs against which measurements could be plotted and assessed. Measurements were taken with the woman in a supine position with legs extended and with the bladder empty, from the top border of the symphysis to the uterine fundus in the midline. No correction was made for fetal presentation, descent of the presenting part, amniotic fluid volume, uterine obliquity, maternal height or weight.

The sensitivity ( the probability of an accurate positive finding ) of fundal height measurement in prediction of low birth weight was greatest (78% ) when compared to abdominal girth ( 55% ) and the ratio ( 33% ) for women considered "high risk". The specificity ( the probability of an accurate negative finding ) was high with both fundal height and the ratio ( 88% )and lower with the abdominal girth measurement ( 75% ).Of 150 women 27% ( 40 ) delivered low birth weight infants.

Fundal height had a specificity and sensitivity of 84%, the abdominal girth measurement had a specificity of 80% and a sensitivity of 58%, and the ratio had a specificity of 93% and a sensitivity of 21% for women considered " low risk". Of 208 women, 9% ( 19 ) delivered low birth weight infants.

Only fundal height with its high sensitivity and specificity appeared to be useful in prediction of low birth weight infants in a comparison with both AG and FH/AG ratios. Fetal growth abnormalities are a global phenomenon and this study compares the symphyseal-fundal height measurement with 2 alternative measurements to determine its usefulness. It was not as influenced by maternal weight and skin fold thickness as abdominal girth measurement and it did not have the unacceptably low sensitivity of the ratio.This study demonstrates that the fundal height measurement can be helpful in prediction of small for gestational age infants particularly in an area where the ultra sound scan might not be available. The standard graphs might only be appropriate to the study population given that ethnic variation is a recognized phenomenon. The study addresses both low and high risk populations to allow a comparison of reliability and usefulness of the measurements considered.

Tape fundal height measurement was compared with an ultrasound technique by Euan, Connor, Hahn, MacMillan and Arheart ( 1995 ). The study compared manual and ultrasound measurement of fundal height. Women were further separated statistically into two groups to determine whether manual measurement is affected by obesity or race.

One hundred and fifty nine women with a singleton pregnancy between 11 and 42 weeks gestation were selected from consecutive obstetric clients presenting to the family practice residency center for routine prenatal care and ultrasound between January and April 1991. Participants were statistically divided into two groups: those with a normal weight for height and those classified as obese.The results were based on 119 women as 40 did not receive both measurements because of differences in medical coverage arrangements.

Ultrasound was used to determine the fundal height and fetal age was then determined by using multiple measurements on all subjects. Fundal height measurements were obtained in a uniform manner with the tape face down to eliminate bias. Two methods were used to identify obesity (ideal body weight and body mass index). Simple linear regression and analysis of covariance were used to analyze the data for correlation between manual and ultrasound measurements of fundal height and differences according to obesity or race respectively.


Results indicated no significant difference between manual and ultrasound measurement of fundal height ( P<.01,R2 of .92). The results of the regression analyses and analysis of covariance demonstrated that obesity had no influence on the relation between ultrasound and manual fundal height measurement for either definition of obesity. The results also demonstrated that race had no influence; both methods of measurement had the same predictive power regardless of race.

When considering the validity of the symphyseal-fundal height measurement, a comparison with ultrasound measurement is clearly valuable particularly as the study suggests that both methods have the same predictive value regardless of obesity or racial group.This study demonstrates that, when carried out using a consistent technique, a tape and a pair of hands prove equally accurate in assessing gestational age as the technology of ultrasound and therefore it supports the validity and, with continuity of technique, the reliability of fundal height measurement.

One of the larger controlled studies to evaluate fundal height measurement was carried out in England by Gardosi and Francis ( 1999 ). The purpose of the study was to evaluate a policy of standard prenatal care which included plotting fundal height measurements on customized prenatal charts as a screening method for fetal growth. The hypothesis was that the study protocol would increase prenatal diagnosis of small for gestational age from 25% to 50%.

A total of 1272 consecutively selected women with a singleton pregnancy and a dating ultrasound performed prior to 22 weeks gestation were selected from 2 defined and separate areas with similar delivery rates and socioeconomic characteristics. Providers for both groups were midwives and G.P.s in Nottingham, England. One group formed the study group (n=667) and the second the control (n=605). Recruitment occurred between May 1994 and March 1995.

The study was prospective, non-randomized, controlled, and population based. A customized prenatal growth chart was produced for the study group which displayed computer generated curves for fetal weight and fundal height adjusted according to physiological characteristics of pregnancy. By means of multiple regression coefficients derived from the population, the optimal weight predicted at term was calculated by adjusting for maternal height, initial weight, parity, ethnic group, and by excluding variables such as smoking. As a result of adjustment for unique physiological characteristics, the slope over time of fundal height varies for each pregnancy. The control group received traditional prenatal care.

The data were analyzed with statistical software ( SPSS for Windows, Version 7.0 ) and spreadsheet ( Microsoft Excel, Version 5.0 ) with add-in functions for odds ratio, confidence limits and P values according to standard formulae. The pregnancy characteristics in the study and control groups showed no significant statistical differences. A significantly higher proportion of SGA infants were suspected in the study group (47.9% vs 29.2% ) yet there were no differences in outcomes between the two groups ( 10.6% and 11.9% respectively were considered SGA or < 10th percentile at delivery ). More LGA infants were suspected in the study group (45.7% vs 24.2% ) and actual outcomes were 12.1% for the study group and 10.2% for the control group. There were fewer referrals of women in the study group for further investigations.The results showed a significant increase in the detection of infants who were SGA or LGA by the strategy adopted in the study. In the control group, the detection rate of SGA was 29% while in the study group 48% of SGA infants were detected.

A study which suggests a method for improving the validity and reliability of fundal height measurement is of undoubted relevance to this study. This research highlights the importance of the fundal height measurement as part of an overall clinical assessment and the value of serial measurement in prediction of fetal growth abnormalities. The precision of the measurement can be improved by using customized growth charts and their use may result in fewer referrals and thus unnecessary investigation and reduced maternal anxiety.

Apart from the fetus as the object of interest, fundal height measurement can also be influenced by surrounding amniotic fluid, placenta, myometrial thickness, abdominal wall fat, and the relationship of the uterus to the bony pelvis (Gardosi & Francis 1999). It is the fetus that is most rapidly changing in size and serial assessment will reduce the influence of the other factors. When fundal height increment was compared retrospectively with measurements on neonates, a highly significant correlation was demonstrated between slow fundal height growth and SGA infants. The true value of fundal height measurement as a predictor of fetal growth must come from serial assessment of change. The researchers concluded that their increased detection rate occurred because of a combination of factors including:

1) The fact that the study group was taught a "proper" uniform method of fundal height measurement.
2) Customized limits for SGA and LGA made the measurement more sensitive.
3) Serial plotting on growth curves can reveal slow growth more clearly.

McGeown ( 2001 ) has also considered fundal height measurement and methods to improve the reliability and validity of the intervention. Consistent failure to recognize SGA fetuses prenatally, despite numerous screening interventions, has led to the development of a growth chart specific to each pregnant woman on which the fundal height measurement can be plotted against individual growth expectations generated from a software package used in conjunction with standardized fundal height measurement. This combination has been found to be an effective, simple and relatively inexpensive method of improving detection. This study considers the effectiveness of current practices and suggests how detection of fetal growth abnormalities might be improved.

The author considers definitions of fetal "growth retardation" and describes how these may not adequately address the issue as they do not always consider individual characteristics such as gestational age, parity or ethnicity. Fetal growth abnormalities are considered in relation to the epidemiological studies which have highlighted an association between "unexplained" stillbirth and IUGR. The author's region (the West Midlands of the UK) has the highest perinatal mortality rate in England and Wales and analysis has suggested that up to 60% of all stillbirths reported were SGA (Trent analysis).This was when compared with a gestational age specific fetal weight standard derived from normal, term live births. This had previously gone unrecognized as it is common practice for pathologists to evaluate stillbirths against a standard derived from the weights of other stillborn babies which may not be valid given the tissue changes that inevitably take place over time.

Current methods of clinical assessment for fetal growth abnormalities are reviewed. Abdominal palpation is the traditional method and the author states " there is no scientific foundation for the use of the umbilicus as a landmark to indicate a specific gestational age, ie umbilicus = 22/24/26 weeks, dependent upon which text one reads"

Fundal height measurement has not proved any more effective than the traditional assessment according to some studies ( Lindhard, 1990 ) and lack of uniformity of method and clinician bias are thought to contribute to the ineffectiveness. Standardization of the technique and use of a growth chart are suggested as a means of improving reliability.

Ultrasound is discussed and, whilst its usefulness is acknowledged, the English National Health Service would not be able to commission serial ultrasounds because of expense. The author also notes that repeated ultrasound would further "over medicalize" potentially normal pregnancies and cause unnecessary stress and inconvenience to families. ( Ultrasound is only available in large hospitals in England and most prenatal care is community based.).

To increase prenatal detection of fetal growth abnormalities, a software package has been developed which generates individualized prenatal growth charts. The "GROW" (Gestation related optimal weight ) program uses computer generated growth charts which display curves for both fundal height and fetal weight. The optimal fetal weight is calculated by adjusting for the physiological variables of maternal height, weight in early pregnancy, parity, ethnicity and excluding known pathological variables such as smoking. The individual calculations are based on coefficients of the significant variables that affected birth weight derived from a stepwise multiple regression analysis of a dataset of 40000 ultrasound dated deliveries.

Standardized fundal height measurements plotted on a customized growth chart significantly increased the prenatal diagnosis of SGA and LGA in the study group. In addition, fewer referrals for further testing were made perhaps as a result of an increase in midwife confidence in their clinical skills and interpretation of fundal height measurement. There was no difference in perinatal outcome between the two groups and thus the rate of true positives was maintained. The article is well written and considers fundal height measurement in relation to reducing the high unexplained stillbirth rate in a specific region in the UK. It discuses a method of enhancing the reliability of fundal height measurement but it does not comment in detail on the number of study participants or provide any specific statistics. It does suggest that further larger studies are necessary to confirm their findings.

The article, despite its lack of specific detail, is very relevant to this study because it suggests a method to improve the validity of fundal height measurement. It embraces the holistic approach so compatible with the philosophy of midwifery and emphasizes the importance of continuity of care.

The "GROW" package is down loadable free from: http//www.gestation.net.

Studies that question the reliability and validity of fundal height measurement.

One of the few prospective randomized controlled trials to evaluate fundal height measurement was not able to demonstrate that the procedure was useful in prediction of SGA infants ( Lindhard, Nielson, Mouritsen, Zachariassen, Sorensen & Roseno,1990 ). The study sought to evaluate the effectiveness of the fundal height measurement in screening for the SGA fetus; the hypothesis being that serial fundal height measurements would more accurately predict the incidence. A total of 1639 women were randomized into a fundal height measurement group and a control group. The fundal height was measured from the 29th week of pregnancy and the measurement was not found to be helpful in the prediction of SGA. There were no significant differences between the two groups regarding the number of interventions, additional procedures or the condition of the neonate at delivery.

All pregnant women attending the prenatal clinic of Gentofte University Hospital, Copenhagen during 1986 were submitted prospectively to the trial at about 14 weeks gestation. Women with uncertain dates, no ultrasound prior to 20 weeks, multiple gestation or having received care elsewhere were excluded. A total of 1639 women participated and their providers were midwives and obstetricians.

On their first visit to the prenatal clinic, women were randomized into 2 groups by drawing a sealed envelope containing a project number- an uneven number resulted in allocation to the fundal height measurement group, and an even number to the control group. In the study group ( n=804) fundal height measurements were taken with a non elastic tape at each prenatal visit from 28 weeks gestation. The results were plotted on a growth chart. In the control group ( n=835) women received fundal height measurement with a tape with no markings and the tape was cut off and not measured until after delivery. The fundal height measurement was obtained in a uniform manner. A growth curve was considered abnormal if it demonstrated a fall of 20% for at least 2 consecutive values or if it remained static for 3 consecutive measurements.

Both groups were comparable in terms of prepregnancy characteristics ( weight, height, parity, age, substance use ) and with regard to pregnancy complication ( polyhydramnios, pregnancy induced hypertension ). Suspicion of IUGR arose in 10% of pregnancies in both groups. For calculation of significance the x2-test was used ( limit of significance : P<0.05 ). There were no significant differences between the 2 groups in respect of diagnostic procedures and interventions due to suspicion of IUGR or any differences in fetal outcome. The hypothesis that the introduction of fundal height measurements would lead to more accurate prediction of SGA infants in the study group was not supported. Infact the study demonstrated better sensitivity and prediction of AGA infants in the control group with masked measurement.

Whilst this study demonstrated that fundal height measurement is of no predictive assistance, the author points out that this does not necessarily mean that it has no diagnostic value in itself. The possibility of clinician bias in the study group is acknowledged. The article is important to this inquiry because it is a large research project which questions the validity of fundal height measurement in the prediction of fetal growth abnormalities. It is interesting to note that guidelines were agreed upon before the trial started but they were clearly not always followed. In 6% of the study group and 5% of the control group, the fundal height was not measured at all!

One of the most prominent midwife researchers to consider fundal height measurement in detail is Janet Engstrom (1993) whose work led to a series of articles published in " The Journal of Nurse-Midwifery". In the first article, a variety of techniques for obtaining fundal height measurements were considered as different measurement techniques can yield varying results. Further studies are advocated to evaluate the 3 measurement techniques to determine if one proves more reliable and valid than others.

Fundal height measurement techniques.

1) Comparison of the height of the uterine fundus to landmarks on the maternal abdomen.
A historical review of this traditional technique is presented with thorough references to as far back as 1752 when Smellie described the relationship between the uterine fundus and maternal abdominal landmarks. Abdominal palpation of the uterine fundus was an important diagnostic aid in a time when diagnosis of pregnancy relied entirely on clinical assessment. The symphysis pubis, umbilicus and xiphisternum were used as reference points and the fundus palpated and described in relation to these landmarks. The symphysis pubis, for example, represents a 12 week gestation uterine size, the umbilicus represents 24 weeks and the xiphistrenum 36 weeks. ( Mayes,1989 ). This technique is still widely used in Europe.

Questions about the reliability of this technique arose as early as 1875 when, as Engstrom describes, Sutugin noted that distances between the anatomic landmarks could vary considerably between women. He found that the distance between the symphysis pubis and the umbilicus, for example, could vary from 15-22cm and this could greatly influence the impression of gestational age. Several investigators subsequently confirmed these initial observations and by the early part of the twentieth century measurement of the fundal height with a tape or pelvimetry caliper was being recommended ( Mc Donald ,1906 ).

2) Tape measure techniques.

Fundal height measurement with a tape became prevalent in the US following publication of McDonald's work in 1906 and 1910. He described a technique where the tape is held with one hand on the upper border of the symphysis pubis, it remains in contact with the skin of the maternal abdomen and extends to the uppermost point of the uterine fundus. Many texts describe fundal height measurement with a tape as a "McDonald measurement".

Variations of this measurement technique have been advocated over the years and Engstrom describes them individually. All begin at the symphysis pubis; one does not include the upper curve of the fundus and another does not involve contact between the maternal skin and the tape. All measurements were obtained in the midline and each variation was an attempt to improve reliability.

3) Caliper measurement techniques.

Caliper fundal height measurement techniques are rarely used in current clinical practice yet some researchers have suggested that they may provide greater reliability ( Engstrom 1989, Varney 1987 ). Originally, the measurement involved placement of one branch of the caliper in the maternal vagina and the other against the upper most point of the fetus in the uterine fundus. Because of maternal discomfort and the risk of infection, the original technique was modified and the clinician's index and middle fingers were placed in the vagina and the distance between a point on the examining hand and the fetal buttocks in the fundus was measured. In the 1950's, when rectal examinations were often performed, measurements were obtained by placing the examining finger on the fetal head through the maternal rectum with the caliper secured to the examining hand with a rubber band !

Comparison of fundal height measurement techniques - some preliminary studies.

Reliability studies.

Four clinicians obtained caliper and tape fundal height measurements from 22 women. No specific instructions regarding technique were provided and interexaminer reliability proved greater when the caliper was used. The small sample size would be considered a limitation to the study buts its finding should justify further research.

Validity studies.

Engstrom studied the validity of fundal height measurement in prediction of birth weight during labor. Two techniques were used ( one that included measurement of the upper curve of the uterus and one that did not ). The measurement technique that did not include the upper curve of the fundus had a slightly higher correlation with birth weight.

In a subsequent study, the predictive value of a fundal height measurement obtained 7 days prior to delivery was explored using 3 different techniques ( 2 as described above and a caliper technique ). In this study, a higher correlation with birth weight occurred in the group where the upper curve of the fundus was included in the measurement. A tape measurement had higher correlation with birth weight than the caliper technique. No sample size is included in the article to know the scope of the study.

In a third study described in the article, the three previously described techniques were compared in a study to consider the relationship between fundal height and gestational weeks. All 3 measurements were obtained for 45 women who delivered AGA infants. The percentage of variance in gestational weeks for the 2 tape techniques were very similar and the caliper method explained less.

These preliminary studies by Engstrom suggested that the caliper method may have greater inter examiner reliability than measurements obtained with a tape but tape measurements have a higher correlation with gestational age and infant birth weight. Engstrom's article is extensive, well written and places fundal height measurement in its historical perspective. The review of the literature is thorough and, whilst the preliminary studies are small, they are important when considering the reliability and validity of fundal height measurement. That various measurement techniques can provide clinically significantly different results is clearly illustrated. The descriptions of the different measurement techniques and their origins are both informative and fascinating. Consistency in technique and conditions is highlighted and the suggestion of a written protocol is made. Other published studies are astutely critiqued for their lack of description of the methods used. The need for additional research in order to reach a conclusion is recognized and randomized clinical trials are advocated.

In the second of her articles, Engstrom (1993) considered the intra and interexaminer reliability of 3 fundal height measurement methods - two tape and one caliper measurement techniques as described by Varney in the midwifery text. Findings indicated that in order to maximize reliability, fundal height measurements should be obtained by the same clinician throughout pregnancy and it is suggested that calipers may be the most reliable method of obtaining the measurement and should therefore be investigated further, although their acceptability to women should also be considered.

A sample of 60 women were selected from private practices where care was provided by both midwives and obstetricians. Subjects were a minimum of 16 weeks gestation and consented to participation. Three nurse-midwives with varied clinical experience and an obstetrician agreed to their measurements being studied and all were instructed in the 3 fundal height measurement techniques.

The 3 fundal height measurement techniques studied were: a tape measure method that included the upper curve of the uterine fundus in the measurement; a second tape technique that did not include the upper curve; and a caliper technique measuring from the symphysis pubis to the uterine fundus. Blank tapes were used so that clinicians were blinded and the numerical markings on the caliper were concealed with tape. Fundal height measurements were obtained in succession using the 3 techniques and each was obtained twice by 2 different examiners. The clinician marked the tape with ink at the appropriate point and, once all of the measurements had been obtained, the data collection material was placed in a sealed envelope until all of the participants had been assessed.

Data was analyzed using the Statistical Analysis System computer program and the statistics presented were: the mean absolute value of the differences between measurements; the standard deviation of the net differences; the technical error of measurement; the % of differences less than or equal to 1cm; and the % of differences less than or equal to 2cm. The demographic, obstetric and anthropometric characteristics of the study subjects were presented. Comparison of the 3 techniques demonstrated that the mean absolute differences for all 4 examiners was smallest for the caliper method ( 0.92 compared to 1.13 and 1.21 for the tape techniques.). Interexaminer differences were very apparent. The largest maximal difference between the measurements of 2 clinicians was 11.5cm for the 'over-the-curve' tape method!

The research is placed in its historical context well and the term 'reliability' used in the title is very clearly defined in the introduction. Several essential findings and suggestions make this article very significant to this study. The importance of continuity is demonstrated. Whilst the caliper measurement might seem superior in terms of reliability, the author does compare their findings to similar studies and suggests that further research is necessary before the use of calipers is recommended in clinical practice. Again, it would seem important to also consider the appeal of the caliper technique to both women and providers.

All of the fundal height measurement techniques considered demonstrated some degree of interexaminer difference and therefore Engstrom provides some specific suggestions for improving reliability:

1) Written protocols should exist that specify exactly how the fundal height should be measured.
2) Landmarks for measurement must be accurately identified.
3) Perhaps the fundal height should be measured more than once at each visit as accuracy of the measurement may be improved with repetition.

Reliability within and between clinicians is vital for safe practice and the differences demonstrated in this study are not really acceptable if fetal growth abnormalities are to be detected and unnecessary interventions avoided, as well as essential interventions and assessments initiated.

In the third of her articles, Engstrom (1993) considered the effect of maternal position on the outcome of fundal height measurement using 4 different positions and the findings confirmed that consistency is essential if fundal height measurement is to be of value in screening for fetal growth abnormalities. The purpose of the study was also to determine which position was preferable to the women and to the clinicians. A sample of 48 subjects was selected with the following criteria: Prepregnancy BMI < 25 controlling for maternal obesity; gestational age between 21 and 36 weeks; and consent to participate. Four midwives were involved in data collection and each obtained fundal height measurements on 48 women for a total sample size of 192.

Women were asked to void within 30 minutes of the examination controlling for bladder distention and clinicians were provided with instruction and descriptions of the 4 positions being studied. These were as follows:

1) Supine.
2) Trunk elevation (15 degrees ) with knees extended.
3) Knee flexion ( 105 degrees )
4) Trunk elevation with knee flexion.

All joint and table angles were measured and the sequence of maternal positions was determined by a fixed randomization. Fundal height measurements were obtained with an unmarked paper tape, controlling for clinician bias, which remained in contact with the skin of the maternal abdomen for the entire length of the uterus. All 4 measurements were placed in a sealed envelope until data collection was completed. The effects of parity, race, and gestational age were also considered. Both the women and the clinicians were invited to comment upon which position they preferred.

Data were analyzed using the Statistical Analysis System and the Statistical Package for the Social Sciences and are presented in 5 tables. The findings demonstrate that maternal position does influence the outcome of fundal height measurement. Even an elevation of 15 degrees resulted in an average difference of 0.95cm. With trunk elevation and knee flexion, the mean difference from measurements obtained in the supine position was 1.69cm. Of the 4 positions considered, supine was preferred by the clinicians because of ease of measurement and, infact, most existing fundal height measurement studies are based on this position. Women, on the other hand, were more comfortable in positions where their knees were flexed and/or their trunk elevated.

This article and its findings are of undoubted relevance to this study and it is important, again, because it is further evidence of the work of the midwife scientist. The importance of continuity has been established and continuity of maternal position would seem to be equally important in obtaining a fundal height measurement that is both reliable and valid An interesting outcome and question raised by this study is that the position for fundal height measurement preferred by women is not that selected by clinicians. Unfortunately, there does not seem to be any data to support a position other than supine at present- another area for further research.

Part Four of Engstrom's articles considered the ability of the clinician to actually identify the uterine fundus. If fundal height measurement is to prove valid then it is vital that clinicians be able to identify the uterine fundus correctly. The purpose of this study was to assess clinicians' ability to identify the uppermost border of the fundus and their findings were compared with real time ultrasound location. Examiner error did not appear to be influenced by variables such as maternal height, prepregnancy weight, weight, parity, gestational age or the amount of fat on the anterior abdominal wall. It was influenced by thickness of the uterine wall and fetal presentation. A sample of 126 pregnant women were selected from those registered for prenatal care at a university hospital. Selection criteria required only that the uterine fundus be palpable abdominally and that an ultrasound was already scheduled for that day. Verbal consent for participation was obtained. Eight clinicians took part (1 obstetrician, 3 CNMs, 2 SNMs and 2 ultrasonographers ).

The clinicians were asked to locate the uppermost border of the uterine fundus and to mark that point with a pen. No specific instructions were provided. For the majority of participants, no position change or significant lapse of time occurred before the ultrasound was carried out. The ultrasonographer used a spinal needle ( the tip covered with an eraser) to create an ultrasound shadow indicating the skin marking, identified the fundus, and measured the distance between the 2 points.

To determine whether maternal physical characteristics influenced the clinician's ability to locate the fundus, the following additional ultrasound assessments were carried out.

1) Assessment of the thickness of maternal fat on the abdominal wall.
2) Assessment of uterine wall thickness.
3) Assessment for the presence of the placenta or fetal parts in the uterine fundus.

Analysis was carried out using the Statistical Analysis System computer programs. An analysis of variance demonstrated that the differences between individual examiner's errors were not statistically significant when considering clinical assessments and ultrasound findings. Examiner error was not significantly influenced by parity or the presence of the placenta or fetal parts in the uterine fundus. Uterine wall thickness did appear to affect examiner error which was greatest when the uterine wall was thin. Fetal presentation also influenced examiner error with a cephalic presentation producing the least error. Maternal characteristics such as height, prepregnancy weight, current weight, prepregnancy BMI, gestational age and anterior abdominal wall thickness produced no significant correlation with examiner error. Clinician error exceeded 1cm in 42.1% of cases and 2cm in 20.6% of cases.

Unfortunately, not all of the clinicians were blinded to the ultrasound results as 2 were also investigators and therefore often present during the ultrasound. This was taken into account in calculations and it was not thought to have influenced outcomes. The sample size was small and not all of the assessments took place under identical conditions- this again was not thought to have affected outcomes.

The article is important to this study because for fundal height measurement to be reliable and valid, it is essential that clinicians be able to identify the uterine fundus accurately. Contrary to the findings of prior studies ( Bailey, 1986 ), this research demonstrates that some maternal anthropometric factors and pregnancy characteristics can contribute to error in fundus identification and thus fundal height measurement. Further studies are warranted.

In the fifth and final article concerning fundal height measurement, Engstrom (1994) sought to determine whether clinician's fundal height measurements are biased by their knowledge of gestational age and their ability to read the numeric markings on a tape measure when measurements are obtained. The researcher successfully places the research problem into a larger theoretical context by considering available literature which note the experimenter expectancy/observer expectancy/observer bias phenomenon. The literature supports the article's hypothesis that clinicians obtain findings consistent with their expectations and, whilst this is not a new theory, an attempt is made to test the hypothesis through quantitative research. A sample of 240 women was selected from women registered for antenatal care at 2 different clinical settings. (A large university hospital and a private midwifery practice). The sample size and source diversity was adequate to be able to draw a conclusion. Women were included between 15 and 40 weeks gestation with a BMI of less than 29.

Marked and unmarked tapes were used to obtain fundal height measurements by physicians, CNMs and student NMs. All measurements were obtained under the same conditions (Eg supine position/legs extended/empty bladder ). The client's record was reviewed prior to measurement. All data materials, including the marked and unmarked tapes, were placed in sealed envelopes and were measured twice by separate investigators to the nearest mm using the same metal ruler.

Data were analyzed to determine whether clinicians' measurements were affected by using a marked tape or by their knowledge of gestational age. A computer program called Statistical Analysis Systems was used and differences between clinicians' measurements were described using MSD (mean absolute differences) and the % of differences less than or equal to 1cm and less than or equal to 2 cm. The results are presented in the narrative concisely and by use of bargraphs and dot plots. There is no suggestion of bias other than that which the research sought to demonstrate!! At 12 pages the article is long but still concise. The abstract and title clearly summarize the topic. The findings confirm that clinicians' fundal height measurements are biased by using a marked tape and by knowledge of gestational age. The article is of undoubted relevance to this inquiry and there is definite correlation between the hypothesis and the methods used to address it. The authors demonstrate through quantitative research that fundal height measurement can be biased, which raises further questions about technique and usefulness. In some ways, the research raised more questions than it answered.

Summary

It has long been recognized that some form of prenatal assessment of fetal growth is necessary in order to be able to detect abnormalities and initiate interventions to improve fetal outcomes. The widespread introduction of fundal height measurement theoretically may improve the detection of those at risk of growth abnormalities but the results have varied widely from studies undertaken at different centers. ( McGeown, 2001 ).

The most recent Cochrane Library review ( Neilson, 2001 ) and the Royal College of Obstetricians and Gynaecologist's clinical audit ( 1997 ) found little evidence to support the use of fundal height measurement in clinical practice. The largest prospective randomized controlled trial undertaken ( Lindhard, 1990 ) concluded that fundal height measurement was not helpful in detection of SGA and that its use resulted in no differences in terms of neonatal outcome. The authors do acknowledge that it would be unwise to abandon the practice until more is known, therefore research is ongoing.

Several maternal, environmental and clinician factors can influence the reliability and validity of fundal height measurement and these have been addressed through research. Maternal factors include uterine wall thickness, the status of the bladder and, according to some studies, race. ( Buhmann ,1998 ). Maternal position and the angle of the head of the table can also affect the result. ( Engstrom, 1993 ). Most influential, ironically, is the affect of clinician continuity, their choice of technique, and clinician bias. ( Engstrom, 1993, Lindhard, 1990 ).

Studies have addressed the issue of improving the technique of fundal height measurement, and thus its reliability, and encouraging results have been shown with incorporation of customized prenatal growth charts. ( Gardosi 1999, McGeown, 2001 ). It is also interesting to note that the reliability of fundal height measurement does not seem to be affected by maternal weight ,height, parity, or gestational age.

The research suggests that the reliability and therefore the validity of fundal height measurement can be improved by ensuring continuity and it is the aim of this study to consider those factors that affect reliability and suggest a protocol to assist clinicians to improve their technique.


Conceptual Framework.

In order to determine the value of a conceptual framework, it is necessary to consider the following components and definitions. A theory is a systematic, abstract explanation of some aspect of reality ( Polit, Beck & Hungler 2001 ) and it is an essential aspect of research. A theory generally comprises of at least 2 concepts that are related in a manner that the theory purports to explain. A conceptual model is concerned with concepts that are linked to a common theme and a conceptual map represents the phenomenon in the form of a diagram to provide a visual outline of the basis of the study. A conceptual framework guides the research process and enables the researcher to link findings to the existing body of knowledge. It is a vital aspect of a study because it will determine the hypothesis and the ultimate study design.

Fundal height measurement is a commonly used clinical intervention to screen for fetal growth abnormalities during the second and third trimesters of pregnancy, thus the framework for this study is physiologic. Because of the anatomic and physiologic changes that take place in the uterus during pregnancy, the fundus becomes evident abdominally and therefore measurable. As it is not feasible to measure the fetus directly at prenatal visits, theoretically the growing fundus becomes a marker variable for fetal growth.

Pregnancy results in profound anatomic and physiological changes to the uterus and these should be reviewed in discussion of the theoretical framework of fundal height measurement. During pregnancy, the uterus increases in weight from 60g to 1000g at term. Its size increases from 7.5cm long by 5cm wide by 2.5cm deep to 30cm long by 23cm wide by 20cm deep. The endometrium thickens and is called the decidua. Growth of the muscular tissue is due not only to an increase in the length and thickness of already existing fibers ( hypertrophy ), but to the development of new fibers ( hyperplasia ). The 3 layers of the myometrium become more clearly defined.

The shape of the uterus changes from the non-pregnant pear shape to a globular form during the first 12 weeks of pregnancy. By 20 weeks of gestation, it becomes ovoid and maintains this shape until term. In the early weeks of the pregnancy, the isthmus increases in length from 7mm to 25mm. By 12 weeks, the developing ovum, now termed the fetus, fills the uterine cavity, the elongated isthmus having expanded to accommodate it. The upper two thirds of the body of the uterus is muscular. In the lower third, the proportion of muscle diminishes and is replaced by connective tissue. The isthmus subsequently develops into the lower uterine segment.

During the first 12 weeks of pregnancy, the uterus enlarges in the cavity of the pelvis. At 12 weeks, it rises into the abdominal cavity and the fundus can be palpated just above the brim of the pelvis. The uterus now loses its position of anteversion and anteflexion,or in some cases retroversion and retroflexion, and becomes erect or tilted slightly to the right (right obliquity ). Abdominal enlargement occurs and may be more prominent when a woman is standing than when she is supine. Abdominal enlargement may be more obvious in multiparas because of loss of muscle tone in the abdominal wall. Approximate expected fundal heights related to maternal abdominal landmarks are as follows:


Gestational weeks. Approximate expected fundal height.
12 Level of the symphysis pubis.
16 Half way between the symphysis pubis and the uterus.
20 1-2 finger breadths below the umbilicus.
24 1-2 fingerbreadths above the umbilicus.
28 -30 1/3 of the way between the uterus and the xiphoid process.
32 2/3 of the way between the uterus and the xiphoid process.
36-38 1 finger breadth below the level of the xiphoid process.
40 2-3 finger breadths below the level of the xiphoid process with lightening.

( Varney, 1997 )

If fundal height measurement is principally aimed at the detection of fetal growth abnormalities, then it is important to consider normal fetal development during the second and third trimesters when the fundal height is measurable. There are 3 phases of cellular growth in the fetus. The initial stage is characterized by cell multiplication ( hyperplasia ). The second stage is characterized by both hyperplasia and hypertrophy and the third by increasing hypertrophy with minimal hyperplasia. Fetal growth is also influenced by genetic factors, availability of nutrition and adequate vascular support. ( McFarlin 1994 ).

The second trimester is 15 weeks long and includes gestational weeks 13-27 by the LMP. Body growth accelerates and there is rapid progress in bone development by week 16. The average crown- rump length at this gestational age is 4.5 inches and the fetal weight is between 3.5 and 4oz. Stronger fetal movement and a thinner uterine wall results in the woman's experience of 'quickening' around the 18th week. By the end of the 20th week the average crown-rump length of the fetus is 6.5 inches and the weight 0.75lb. The average crown-rump length at 24 weeks gestation is 8 inches and the weight is approximately 1.25lb.

The 3rd trimester is 13 weeks long and includes gestational weeks 28-40. A substantial weight gain takes place at week 28 and the fetal length is approximately 9 inches and the average weight is 2.25lb. By 32 weeks, subcutaneous fat deposits contribute to an average weight of 3.75lb and the crown-rump length reaches 11 inches. At 36 weeks gestation, the fetal weight is approximately 5.5lb and the crown-rump length is 12.5 inches. By term, the crown-rump length averages 14 inches and the fetal weight 7.5lb.( Sweet ,1988 ).

Conceptual Map.

FHM + MF + FF + EF + CF = +/- DFGA.

FHM = Fundal height measurement.
MF = Maternal factors.
FF = Fetal factors.
EF = Environmental factors.
CF = Clinician factors.
DFGA = Detection of fetal growth abnormalities.

Having reviewed the uterine anatomy of pregnancy and fetal growth, theoretically there should be a correlation between fetal size, gestational age and the fundal height measurement during pregnancy. Research has shown that the accuracy of the fundal height measurement, and therefore its validity, can be influenced by a variety of fetal, maternal, environmental and clinician variables.Theoretically, the ability of the fundal height measurement to detect fetal growth abnormalities could be influenced by maternal factors such as race, parity, weight, height and abdominal wall thickness. ( Engstrom, 1993 ) Fetal factors which may render the fundal height measurement less reliable include presentation, position, multiple gestation, amniotic fluid volume and placental location. ( McGeown ,2001 )The conditions under which the fundal height measurement is obtained may significantly influence the outcome-the status of the maternal bladder and her position could render the intervention invalid. ( Engstrom ,1993 ) Ironically, the clinician herself can contribute to the question of whether fundal height measurement is reliable-clinician bias and error in identification of the uterine fundus can influence the result as can lack of continuity of provider and the actual measurement technique used. ( Engstrom, 1993 ).

Reliability is an essential feature of any instrument used in research or clinical practice ( Polit, Beck & Hungler, 2001 ). Reliability is defined as the repeatability, reproducibility, or consistency of a measurement, when the measurements are obtained under identical conditions. Validity is the extent to which an instrument measures what it is supposed to be measuring. Reliability and validity are not totally independent concepts - a measuring device that is not reliable can not be valid, however an instrument can be reliable without being valid. It is the aim of this study to determine whether fundal height measurement does enable the clinician to identify fetal growth abnormalities or whether the intervention is flawed and, if so, how it can be improved to ensure reliability and validity.

Chapter 3.

The clinical issue addressed by the EB-CPG.

Fundal height measurement is a widely used clinical intervention and even though it is appealing for clinical use because it is quick to perform, inexpensive and relatively non-invasive, research suggests that its reliability and validity may be negatively influenced by the technique used. The accuracy of the measurement, for example, can be affected by maternal position, the status of the maternal bladder, clinician training and the type of measurement technique used. In order to promote continuity and uniformity of technique, an EB-CPG has been produced to improve the reliability and validity of this frequently used prenatal procedure.

Method of development.

The principal source of initial references was a database search of the Midwives Information and Resource Service ( MIDIRS ). "Fundal height measurement" produced 36 references of which 24 were considered relevant. Abstracts were requested and then 18 full text articles were obtained from MIDIRS and the libraries of Philadelphia and East Carolina universities from a variety of journals. Further searches of MEDLINE via "proquest' and "orvis" enabled evaluation of a further 10 abstracts. Midwifery texts ( Varney's Midwifery and Maye's Midwifery ) and obstetric texts ( Williams' Obstetrics and Obstetrics Normal and Problem Pregnancies ) were consulted.

Research which both supported and failed to demonstrate the value of fundal height measurement were considered in the development of the EB-CPG. Descriptive, prospective and randomized controlled trials were included and various methods of fundal height measurement compared.

Evidence considered.

Study authors	Grade of Evidence A=true experiment B= quasi- experiment C= correlation D= description	Study design	Sample size and description	Research procedure	Measures used & their reliability & validity	Statistics reported (include type of statistic reported & p value/confidence intervals as indicated)	Study results & midwifery perspective
Mathews,P. 1987.	B	Prospective.	3 abd measurement methods compared.584 sets of observations obtained from 250 subjects.	Fundal ht (FH)was compared with abd girth (AG) and abd length(ALR) measurements to assess reliability of each in prediction of SGA.	A uniform technique using a non elastic tape was used. FH=78% sensitivity AG=55% " ALR=33% " FH=88% specificity AG=75% " ALR=88% " Only FH appeared useful in prediction of SGA.	Mean and 1&2 SD above and below were calculated for each measurement technique.	Fundal ht demonstrate-d greater reliability, sensitivity and specificity in prediction of SGA. No midwifery perspecive is presented.
Euan,D. et al. 1995.	B	Prospective.	119 women were statistically divided into 2 groups: normal wt and obese by BMI/IBW. All subjects received tape and US FHM to compare the reliability of each technique.	U/S and a tape FHM(face down to eliminate bias) technique were carried out on pregnant women between 11 and 42 weeks gestation.	FHM was obtained using a uniform technique.U/S was used to locate the fundus and the 2 findings were compared. The comparison enabled researchers to support the hypothesis that a manual technique is as reliable and valid as the U/S method of location and measurement of the FH.Both the reliability & validity of FHM were improved by use of the customized individual growth chart.	Simple linear regression and analysis of covariance. P<.01, R2=.92	There was no significant difference between manual and U/S measurement of FH. The reliability and validity of the technique is supported. A midwifery perspective is not represented. Obesity had no influence on either measurement.
Gardosi,J. Francis, A. 1999.	A	Prospective. Non-randomized. Controlled.	1272 participants were selected from 2 separate geographical areas to determine the reliability and validity of FHM & the use of a customized antenatal growth chart.The hypothesis was that the study technique would lead to an increased prenatal diagnosis of SGA.	A control group(n=605) received traditional PNC & the study group (n=667)underwent FHMs which were plotted on customized growth charts adjusted for variables including maternal ht, wt, parity & ethnicity.	By means of multiple regression coefficients derived from the population, the optimal fetal wt predicted at term was calculated by adjusting for maternal ht, wt, parity & ethnicity.	Odds ratio; confidence limits; P values according to standard formulae. SGA= OR=2.23 CI=1.12-4.45P=0.03. LGA= OR=2.63 CI=1.27-5.45 P=0.01	A significantly higher proportion of SGA & LGA infants were detected in the study group.A midwifery perpective is not represented.
Linhard,A. et al. 1990.	A	Prospective. Randomized. Controlled.	1639 participants randomized into a FHM group & a control group to determine the implications of introducing FHM into widespread clinical practice.	The study group (n=804)received FHM from 28 weeks gestation & the results were plotted. The control group (n=835) had the FH measured with a tape with no markings in a uniform position & the results were not analyzed until after delivery.	The hypothesis that FHM would lead to more accurate prediction of SGA infants meant that delivery outcomes were measured. Maternal characteristics were also considered ( age, wt, age, parity, smoking & alcohol)	X2-test ( limit of significance: P<0.05), confidence interval.	Reliability and validity of FHM NOT supported as there were no differences in outcomes between the 2 groups.Midwives were involved in data collection but a specific perspective was not presented.
Engstrom, J. 1993.	B B B B B B B	Prospective. Prospective. Prospective. Prospective. Prospective. Prospective. Prospective. Controlled	22 women received FHM from 4 clinicians to determine the reliability of 3 different FHM techniques. ( Very limited sample size.) 44 women received the 3 FHM techniques to determine which might more accurately predict birth weight. 45 women received FHM using the 3 techniques to consider the relationship between FHM & gestational age. 60 women received FHM using 3 techniques from 4providers to compare inter & intra-examiner reliability. 48 women underwent FHM in 4 different positions to determine the effect on the reliability/validity ( Total 192 measurements). 126 women had the uterine fundus identified by a clinician and then by U/S to determine clinicians' ability to locate the fundus.There were 8 clinicians involved. 240 subjects between 15-40 weeks had FHM obtained using marked & unmarked tape to assess the effect of clinician bias.	No specific instructions were given to clinicians regarding the techniques.( 2 tape methods & 1 caliper method) The 3 FHM techniques were obtained 7 days prior to the EDD & compared in terms of correlation with birth weight. All 3 techniques were obtained from women who delivered AGA infants. Clinicians were instructed in a uniform technique and blank tapes were used to eliminate bias . Between 21-36 weeks gestation FHM were obtained in these maternal positions. 1) Supine. 2)Trunk elevation/knees ext. 3)Supine/knees flexed. 4)Trunk elevation/knees flexed. The uppermost border of the fundus was marked by a clinician following palpation ( no instruction) & then U/S was used to identify it sonographically. FHM were obtained using marked & unmarked tapes & a uniform technique. The client's record was reviewed prior to measurement.	1) FHM tape technique that includes the upper curve of the fundus in the measurement. 2)FHM tape technique that did not include the upper border. 3)A FHM technique utilizing a caliper. As above.A tape technique demonstrated greater validity. As above. As above. Reliability proved greatest with the caliper method. A uniform tape measurement technique was used by 4 different clinicians on each subject using blank tapes. Even a slight variation in maternal position can effect reliability & validity. Validity was influenced by uterine wall thickness(error=greatest when the uterine wall was thin) Inter examiner error was not statistically significant.Validity was not influenced by parity. Reliability was influenced by fetal presentation. Validity of FHM is influenced by clinician bias which is demonstrated by this research.	Mean: Standard Deviation. Correlation. Analysis of variance. SD of net differences. % of differences less than & equal to 1cm & 2cm. Mean absolute differences. Mean; SD; t-value; P-value. Analysis of variance. Mean absolute differences.	The caliper FHM technique demonstrated greater inter-examiner reliability. A higher correlation with birth wt occurred with FHM#1. The 2 tape techniques showed a similar % of varience.The caliper method explained less. The MAD was smallest for the caliper method. Inter-examiner differences were very apparent. When there is trunk elevation & knee flexion the differences between the supine position were clinically & statistically significant. Error in identification of the uterine fundus can be influenced by maternal & fetal factors. FHM reliability & validity is negatively affected by clinician bias. Each of these articles by Engstrom presents the midwifery perspective and midwives were directly involved in each of the studies

Study	Grade of Evidence	Study Design and Validity -Cohort study -Case Control -Case series -Case reports -Expert opinion	Statistics Reported -Odds Ratio -Relative Risk -Confidence Interval -P value	Conclusions/Opinions
Detecting fetal growth abnormalities. McGeown, P. ( 2001 )	11-2	Case Reports.	None specific reported in detail.	Standardizing FHM & routinely serially plotting the measurement on a customized growth chart has been shown to improve the detection rate of fetal growth abnormalities.
Cochrane Data Base. Neilson, J. P. ( 2001 )	1-C	Expert opinion.	Review of existing research.	There is insufficient data to assess whether the routine use of FHM during prenatal visits improves pregnancy outcomes.

Criteria established for effectiveness.

Fundal height measurement has been evaluated as a method of identifying fetal growth abnormalities by a wide range of researchers and, whilst not all reached the same conclusions, there is a consensus that additional studies are warranted. The clinical perspective of the midwife is well represented as a result of the extensive work by Engstrom, a research midwife who has explored the issue from a number of angles, and by McGeown, an English research midwife. Fundal height measurement is an integral aspect of antenatal encounters carried out by midwives and the profession is at the forefront of prior and ongoing studies.

Obstetricians have contributed much to the body of knowledge on the role of fundal height measurement often in collaboration with their midwife colleagues.( Lindhard, 1990, Gardosi, 1999 ) Family physicians continue to provide obstetric care in many centers and their contribution in terms of relevant research is important and thus included.(Euan et al, 1995 )

The perspective of the pregnant woman herself is also represented in an important study conducted by midwife researchers( Engstrom et al, 1993 ).

Population of interest/exceptions.

The population of interest for an EB-CPG concerning fundal height measurement would be all pregnant women in their second and third trimesters of pregnancy when the fundus is measurable and thus acts as a marker variable for fetal growth. There seems to be a general consensus that FHM should commence at twenty weeks gestation. Even where a fetal growth or amniotic fluid disorder is detected ( IUGR, macrosomia, oligohydramnios ) FHM performed serially in a consistent manner can assist in close monitoring of the clinical situation.

All clinicians involved in the provision of prenatal care can benefit from an EB-CPG which emphasizes the importance of continuity of technique and thus improves reliability and validity.

Exceptions to the guideline might include any situation where the validity of the technique might be affected by the clinical condition such as a multiple pregnancy, extreme morbid obesity or polyhydramnios. The use of FHM may be helpful in initial identification of the issue and may used as part of the overall monitoring plan.

Synthesis of evidence and conclusions.

Fundal height measurement has become an integral aspect of prenatal care. Research has demonstrated that both the reliability and validity can be improved by ensuring continuity and uniformity of technique. The ability of the clinician to identify the uterine fundus itself has been evaluated and, even without instruction on technique, the degree of error when compared with ultrasound identification was not influenced by maternal characteristics such as height, prepregnancy weight, current weight, gestational age, race or anterior abdominal wall thickness.( Engstrom, 1993, Euan, 1995 ) Uterine wall thickness and fetal presentation can negatively influence examiner error but, on the whole, fundal height measurement compared well with an ultrasound technique and thus its validity is supported.

A variety of fundal height measurement techniques have been used and evaluated. ( Varney, 1997, Engstrom, 1993.) A caliper technique did demonstrate greater interexaminer reliability but it did not prove popular in clinical practice with either the providers or the women themselves. The fundal height measurement technique which included the upper curve of the uterus in the measurement demonstrated greater interexaminer reliability than the alternative technique that did not include the upper curve ( Engstrom, 1993 ) and this will be reflected in the EB-CPG.

Maternal position during fundal height measurement has been shown to significantly influence the outcome regardless of parity, race, gestational age or the experience of the clinician. ( Engstrom 1993 ) Historically, fundal height measurements have been obtained with the client in the supine position and current knowledge about the reliability and validity is based entirely on this position. Standardizing maternal position is necessary if clinicians plan to use fundal height measurement to make clinical management decisions. Even a small variation from the supine position, such as an elevation of the head of the examination table by 15 degrees, resulted in an average difference in measurements of 0.95 cm in one study ( Engstrom, 1993 ) which is important if clinicians usually expect the fundal height to grow approximately 1 cm per week.

The effect of maternal bladder volume on fundal height measurement has been considered ( Engstrom 1989 ) and pre-void fundal height measurements were significantly larger than post-void ( mean 0.63 cm, SD 1.26 ) The findings indicate that women should be instructed to void within 30 minutes prior to the measurement being obtained.

Clinician bias has been observed in several studies ( Engstrom, 1993, Lindhard, 1990 ) but methods of minimizing the phenomenon have yet to be researched. Perhaps such bias can be reduced by keeping clinicians blinded to their own fundal height measurements during the procedure by suggesting simply turning the tape over while the measurement is being obtained.

The use of customized prenatal growth charts in conjunction with fundal height measurement may result in improved detection of fetal growth abnormalities and fewer referrals for additional interventions. ( Gardosi, 1999, McGeown, 2001 ) Their use should be further evaluated and considered for use in clinical practice.

Of all the factors that may affect the reliability of fundal height measurement, perhaps lack of continuity is the most significant. There is consensus among researchers that continuity of provider and technique are vital to the effective use of fundal height measurement in clinical practice.

Summary statement of findings.

Fundal height measurement is an inexpensive, easily obtainable, relatively non-invasive prenatal intervention that, if provided with continuity, can provide reassurance of fetal growth and contribute to early detection of fetal growth abnormalities.

Outline of an EB-CPG / clinical algorithm.

• Communicate the guideline to all clinicians involved in prenatal care to promote uniformity and continuity.
• Aim for prenatal care to commence during the first trimester of pregnancy.
• Obtain detailed menstrual, obstetric and medical histories to estimate gestational age.
• Physical examination including bimanual pelvic exam to compare uterine size with estimated gestational age.
• Begin serial fundal height measurement at 20 weeks gestation.
• Method : Expose the woman's abdomen sufficiently to allow accurate identification of the abdominal landmarks. Place the zero line of a tape measure face down on the superior border of the symphysis pubis. In the midline, extend the tape across the contour of the abdomen to the upper border of the uterine fundus. Note the measurement in cm and record on the client's chart.
• Maternal position : supine ( head elevated no more than 30 degrees ).
• Maternal bladder : emptied within the preceding 30 minutes.
• Consider use of a customized prenatal growth chart adjusted for individual maternal physical characteristics ( An example may be obtained via http/www.gestation.net ).
• If FHM = EGA : Continue serial FHM at each visit throughout the second and third trimesters of pregnancy ensuring continuity of technique and , if possible, provider.
• If FHM does not = EGA : If a discrepancy is found at the initial visit or if the gestational age can not be reliably determined obtain an ultrasound to assess the gestational age or rule out multiple pregnancy / fetal growth abnormality.

A discrepancy of >2cm is considered significant. If found :
a) Check and confirm dating criteria.
b) Recheck the measurement.
c) Ask the client to return in 1-2 weeks for a measurement recheck and, if a discrepancy still exists, order an
ultrasound to rule out fetal growth abnormality or amniotic fluid disorder.
d) If no abnormalities are detected no further intervention is necessary unless the discrepancy continues to increase.
Serial FHM should continue.
e) If an abnormality is detected follow up, consult or refer as appropriate.

Evaluation / Measurement strategy.

Research exploring FHM produced evidence which both supported and failed to demonstrate the value of the intervention in routine clinical practice. There was, however, a consensus among researchers regarding the need for further studies to evaluate the reliability and validity of FHM. Additional research on the use of customized prenatal growth charts ( Gardosi & Francis, 1999, McGeown, 2001 ) and further work on maternal position, the measurement technique itself and methods of eliminating clinician bias ( Engstrom, 1993 ) have been proposed. So while much has been learned, FHM is an ongoing research area and this is important because any clinical intervention must be substantiated by research.

On a more local level on going evaluation is also vital. FHM is a useful tool for assessing fetal growth only if its reliability and validity are ensured by providing continuity and uniformity of technique. An important aspect of evidence based practice is evaluation of interventions for effectiveness. With delivery of the neonate, one is provided with a measurable outcome. Birth weights can be used to assess the ability of FHM to identify fetal growth abnormalities.


References

American College of Obstetricians and Gynecologists. (1989). Standards for obstetric-gynecologic services. (7th ed.).Washington, DC:Author.

Bailey, S., Sarmandal, P., Grant, J., (1989 ) A comparison of three methods of assessing inter-observer variation applied to measurement of the symphysis-fundal height. British Journal of obstetrics and Gynaecology, 96, 1266-1271.

Benbow, A., Semple, D., Maresh M. (1997 ) Effective procedures in maternity care suitable for audit. Royal College of Obstetricians and Gynaecologists:Clinical Audit, 56.

Buhmann, L., Elder, W., Hendricks, B., Rahn, K.(1998 ) A comparison of Caucasian and south Asian hmong uterine fundal height during pregnancy. Acta Obstetricia et Gynecologica Scandinavica, 77, 521-526.

Crosby, M., Engstrom, J., (1989 ) Inter-examiner reliability in fundal height measurement. Midwives'Chronicle, 102, 254-256.

Engstrom, J.L., Ostrenga, K.G., Plass, R.V. ( 1989 ) The effects of maternal bladder volume on fundal height measurement. British Journal of Obstetrics and Gynaecology, 98, 8, 987-991.

Engtrom, J.L. (1993). Fundal height measurement Part 1.Techniques for measuring fundal height. Journal of Nurse-Midwifery, 38, 5-16.

Engstrom, J.L. (1993). Fundal height measurement Part 2. Intra and interexaminer reliability of three measurement techniques. Journal of Nurse-Midwifery, 38 ,17-27.

Engstrom, J., Piscioneri, L., Low, L., McShane, H., McFarlin, B. ( 1993 ) Fundal height measurement . Part 3. The effect of maternal position on fundal height measurements. Journal of Nurse-Midwifery, 38, 1, 23-27.

Engstrom, J.L. (1994). Fundal height measurement Part 4. The effect of clinician bias on fundal height measurement.Journal of Nurse-Midwifery, 39, 3, 130-141.

Engstrom, J., Sittler, C., Swift, K. (1994) Fundal height measurement. Part 5. The effect of clinician bias on fundal height measurements. Journal of Nurse-Midwifery, 39, 130-141.

Euans, D.,Connor, P.D., Hahn, R., MacMillan, R., Arheart, K. (1995 ) A comparison of manual and ultrasound measurement of fundal height.The Journal of Family Practice,40 (3), 233-236.

Gabbe, S.,Niebyl, J., Simpson, J. (Eds.).(1996 ).Obstetrics. Normal and problem pregnancies. New York: Churchill Livingstone.

Gardosi, J., Frances A. (1999). Controlled trial of fundal height measurement plotted on customized antenatal growth charts. British Journal of Obstetrics and Gynaecology, 106, 309-317.

Lindhard, A., Nielsen, P., Mouritsen, L., Zachariassen, A., Sorensen, H., Roseno, H. (1990). The implications of introducing the synphyseal-fundal height measurement. A prospective randomized controlled trial. British Journal of Obstetrics and Gynaecology, 97, 675-680.

Mathews, M., Prabha, J., Mathurathnam, S.(1987). Screening for light-for-gestational age infants: a comparison of three simple measures. British Journal of Obstetrics and Gynaecology, 94, 217-221.

McFarlin,B.(1994). Intrauterine growth retardation. Etiology, diagnosis, and management. Journal of Nurse-Midwifery, 39, 52-65.

McGeown, P.(2001).Detection of fetal growth abnormalities.Midirs Midwifery Digest,11,190-193.

Neilson, J.(2001). Synphysis-fundal height measurement in pregnancy, Cochrane Review. Cochrane Review. Oxford: update software.

Polit, D., Beck, C., Hungler, B., (2001).Essentials of nursing research. Methods, appraisal, and utilization.(5th ed.).Philadelphia: Lippincott.

Public Health Service. (1989). Caring for our future:the content of prenatal care.Washington, DC:PHS, Department of Health and Human Services.

Rooks, P. ( 1999 ) Evidence-based practice and its application to childbirth care for low-risk women. Journal of Nurse-Midwifery, 44,4, 355-369.

Sweet, B. (1988 ). Mayes' Midwifery ( 11th ed.) London: Balliere Tindall.

Varney, H.(1997).Varney's Midwifery (3rd ed.). Boston:Blackwell Scientific Publications.

. . . . . . .