A Second-Stage Partogram

A. R. SIZER, MB, BCh, PhD, J. EVANS, MB, BCh, S. M. BAILEY, MB ChB, AND J. WIENER, MB, BCh
Objective: To describe a second-stage partogram based on a system of scoring the descent and position of the fetal head and to use this system for studying progress in the second stage of labor and predicting mode of delivery and obstetric outcome. Methods: A prospective observational study of 1413 women at term with a singleton, cephalic presentation. The position and station of the fetal head were observed and scored at diagnosis of the second stage of labor, 1 hour later, and then at 30 minute intervals until delivery was achieved. The score at diagnosis of the second stage of labor was assessed for its ability to predict eventual mode of delivery and duration of labor. A normogram was dened for nulliparas and multiparas and was used to dene normal and abnormal progress in the second stage, associated factors in the rst stage of labor, and mode of delivery. Results: Increasing total score at the start of the second stage of labor is associated with increasing chance of spontaneous vaginal delivery (odds ratio [OR] 1.68 for nulliparas, 1.59 for multiparas), decreasing chance of instrumental vaginal delivery (OR 0.67 for nulliparas, 0.64 for multiparas), and emergency cesarean delivery (OR 0.39 for nulliparas). Abnormal progress as dened by the normogram is associated with use of epidural anesthesia, induction of labor, augmentation, dystocia, and increased incidence of operative delivery. No signicant difference is found between normal and abnormal second stages of labor in fetal outcome as determined by Apgar scores. Conclusion: The second-stage partogram offers an objective basis for management of the second stage of labor. (Obstet Gynecol 2000;96:678 83. 2000 by The American College of Obstetricians and Gynecologists.)

Graphic assessment of progress in labor was rst suggested by Friedman.13 He proposed that of the major observable clinical factors in the rst stage of labor, only cervical dilatation and descent of the presenting part were helpful in the assessment of progress, and he
From the Department of Obstetrics and Gynecology, University Hospital of Wales, Cardiff, United Kingdom. The authors thank J. F. Pearson for the original idea and the Department of Medical Computing and Statistics, University Hospital of Wales, for statistical advice.

introduced the classic curves of mean dilatation1 and descent,3 plotted in a linear fashion, against time in labor. The establishment of a normal curve allowed various aberrations of dilatation and descent to be diagnosed and studied retrospectively.2,4 This approach was developed clinically as a composite partogram,5 with later modications (Studd JWW, Duignan N. Graphic records in labour [letter]. BMJ 1972;4:426).6 The latter version was based on a selected ideal population of nulliparous and multiparous women and included more information on the fetal and maternal condition. However, the original concept of assessing progress in labor graphically by serial examinations of cervical dilatation and descent of the presenting part was retained. Experience in the use of the partogram showed that it claried the recording and identication of abnormalities by comparison with an ideal prole of progress. It was time saving provided that the information was not duplicated, it facilitated teaching, and it allowed the use of normograms and improved recognition of abnormal patterns of dilatation and descent. Before introduction of the partogram, its validity was demonstrated by assessing the ability of a normogram to separate normal labors from those with an abnormal outcome.6 Once full dilatation is reached, the partogram stops and information is not graphically represented, thus losing the advantages outlined above. However, the need for a graphic display of progress into the second stage of labor may continue, especially when the second stage is prolonged, as increasingly occurs with use of epidural analgesia.7,8 Although descent continues, cervical dilation is no longer useful in the second stage, and additional variables are necessary to assess progress. The rate of descent of the presenting part, which may be ascertained by the station of the presenting part relative to the ischial spines, increases in the last part of the active phase of the rst stage of labor. This increased rate of descent continues through the second stage of labor.2 Descent, which has been shown to be an important factor in assessing the progress of the

678 0029-7844/00/$20.00
PII S0029-7844(00)00981-9

Obstetrics & Gynecology

rst stage, may be proposed as a useful tool in the assessment of the second stage. Descent in the second stage of labor is accompanied by rotation of the presenting part as it negotiates the pelvis. Friedman and Sachtleben4 showed that arrest of descent was frequently associated with fetal malpositions occipitoposterior and occipitotransverse positionsand suggested that abnormalities of rotation were important prognostic factors in the second stage. Position is thus a second variable that may be used in observation of progress in the second stage. The aims of our study were to introduce the concept of a second-stage partogram and to see whether it could be displayed easily and usefully in the clinical situation. The partogram was based on a scoring system of the variables of position and station; these variables and their combined score were assessed for their ability to predict eventual mode of delivery. A normogram line was constructed for the partogram on the basis of median scores at each time of assessment in the second stage, allowing classication of progress of the second stage as normal or abnormal. The development of the normogram line allowed comparison with time spent in the second stage of labor as an independent predictor of mode of delivery. Finally, the concept of normal and abnormal second stages was used to examine the effect of associated rst-stage factors (use of epidural anesthesia, induction of labor, dystocia, prolonged 710 interval [interval between 7 cm of dilatation and full dilatation in the rst stage of labor lasting longer than 3 hours9] and augmentation), mode of delivery, and fetal outcome.

Materials and Methods

The study was performed prospectively over 20 months. We enrolled consecutive women delivering in a teaching hospital that performs 3000 deliveries per year. Ethical approval was given by the appropriate authority. All women who reached full dilatation with a singleton pregnancy at term (more than 37 weeks gestation) and a cephalic presentation were recruited by the midwife who attended them throughout their labor. We excluded women who did not attain full dilatation, those who had breech presentation, and those who had multiple pregnancy. A total of 2993 women were eligible to enter the study, and 1413 consented to do so. Participation rates were 58.9% for nulliparas and 38.6% for multiparas. As is normal practice in the United Kingdom, all midwifery staff were involved and were responsible for the conduct of the labor, unless medical obstetric intervention was indicated. Details of the progress of the rst stage of labor, including such complicating factors as administration

of oxytocin for induction or augmentation, use of epidural block, presence or absence of rst-stage dystocia (dened as progress of less than 1 cm of cervical dilatation per hour after 3 cm had been achieved), and a 710 interval longer than 3 hours,9 were noted. Once established labor was diagnosed, vaginal examinations were routinely performed at 2-hour intervals throughout the rst stage and were recorded on a standard partogram.6 The onset of the second stage of labor was diagnosed by vaginal examination or by the clinical nding of a visible vertex. The station and position of the head were recorded. Further vaginal assessment of station and position was performed after 1 hour and then, if possible and if the patient agreed to it, at 30-minute intervals until delivery. The time of delivery was recorded and the length of the second stage was calculated. The method of delivery and indications for any intervention were recorded. No women in the study underwent elective low forceps delivery for medical reasons, such as cardiac disease. No particular time limit was set for the second stage, but it seldom lasted more than 3 hours. Birth weight was noted. All vaginal examinations, recordings, and scoring were performed by the midwives and obstetric staff supervising the labor. Information on station and position was scored as follows: occipitoanterior position was considered the most favorable and was scored 2; occipitotransverse positions were considered less favorablethey may be physiologic or may represent partial rotation of the fetal head to either occipitoanterior or occipitoposterior positionand were therefore scored 1. Occipitoposterior positions were deemed to be the least favorable and to represent malpresentation; these positions were scored zero. Station was also scored. Station higher than 1 cm below the ischial spines was scored zero, station at spines 1 was scored 1, and any station lower than this was scored 2. The maximum total score that could be obtained from position and station was 4. If the vertex was visible and anal dilatation was present, indicating exit from the bony pelvis and imminent delivery (equivalent to spines 4), a score of 5 was awarded. Once delivery had occurred, a score of 6 was assigned, indicating the completion of the second stage of labor. A score was allocated after delivery so that the secondstage partogram could be completed graphically. The sum of descent and position scores was plotted against time elapsed in the second stage to give an indication of progress in each individual labor. The relative importance of position, station, and total score was assessed for its ability to predict mode of delivery by using logistic regression.

VOL. 96, NO. 5, PART 1, NOVEMBER 2000

Sizer et al

Second-Stage Partogram 679

Figure 1. Second-stage partograms with normogram lines (bold line) for nulliparas and multiparas. The panel at left shows abnormal progress in a nullipara (light line with X) that resulted in a forceps delivery 90 minutes after diagnosis of the second stage. The panel at right shows rapid (normal) progress in a multipara (light line with X) that culminated in spontaneous vaginal delivery.

A normogram was constructed for progress in the second stage by taking the median score at each time point of vaginal examination. Only scores less than 5 at the time of diagnosis of the second stage were used to construct the normogram. A normogram was constructed for both nulliparas and multiparas. Progress in the second stage was then categorized as normal or abnormal on the basis of the normogram. Second-stage labors that progressed on or to the left of the normogram line (Figure 1) were classied as normal. Secondstage labors that had any point to the right of the normogram line were classied as abnormal. Progress according to the normogram was used to evaluate mode of delivery, obstetric outcome, and factors in the rst stage of labor that may affect progress in the second stage. These data were analyzed by using the 2 test.

Results
The study population consisted of 744 nulliparous women and 669 multiparous women. Group-specic demographic details are shown in Table 1. The scores for station and position at diagnosis of the second stage of labor are predictive of mode of delivery.
Table 1. Group Demographic Data
Variable No. of women Median birth weight (kg) Use of epidural anesthesia (%) Need for induction of labor (%) Need for augmentation (%) Primary arrest (%) Long 710 interval (%) Median duration of second stage (min) SVD (%) IVD (%) CD (%) Nulliparas 744 3.4 37.5 11.6 25.4 29.0 17.8 67.5 59.2 39.0 1.8 Multiparas 669 3.4 13.6 8.4 11.0 8.4 4.5 14.0 90.6 9.0 0.4

CD emergency cesarean delivery; IVD instrumental vaginal delivery; SVD spontaneous vaginal delivery.

The higher the initial score of either of these variables, the greater the incidence of spontaneous vaginal delivery and the lower the incidence of instrumental vaginal delivery and emergency cesarean delivery (Table 2). However, total score improves the predictive range (the range of separation of incidence of a particular mode of delivery by highest and lowest scores) of mode of delivery to a greater expense than either position or station alone (Table 2, Figure 2). For example, at the time of commencement of the second stage of labor in nulliparas, a station score of 0 (the lowest score) predicts a spontaneous vaginal delivery rate of 36.7%, whereas a score of 2 (the highest score) predicts a spontaneous vaginal delivery rate of 62.3%. Compared with total score, the lowest and highest scores (0 and 5) predict spontaneous vaginal delivery rates of 25.0% and 96.9%, respectively. Total score therefore improves the predictive range. Total score at the start of the second stage is also predictive of time spent in the second stage. In nulliparas, the median time spent in the second stage is 131 minutes at a score of 0, 128.5 minutes at a score of 1, 92 minutes at a score of 2, 78 minutes at a score of 3, 67.5 minutes at a score of 4, and 25 minutes at a score of 5. In multiparas, median time in the second stage is 109.5 minutes at a score of 0, 72.5 minutes at a score of 1, 57 minutes at a score of 2, 37 minutes at a score of 3, 26 minutes at a score of 4, and 10 minutes at a score of 5. The scoring system that we devised was assessed statistically for its ability to predict mode of delivery. Table 2 shows that when total score is entered into univariate logistic regression, each incremental rise in score has a statistically signicant effect on mode of delivery. When the components of position and station at start of the second stage are entered into multivariate logistic regression, each has an independent statistically signicant effect on mode of delivery. Station is not a signicant predictor of cesarean delivery in nulliparas, and station is not signicantly associated with either spontaneous vaginal delivery or instrumental vaginal delivery in multiparas. Odds ratios (ORs) for position

680 Sizer et al

Second-Stage Partogram

Obstetrics & Gynecology

Table 2. Logistic Regression Analysis for Total Score, Position, and Station at the Start of the Second Stage of Labor With Each Mode of Delivery as the Dependent Variable
Mode of delivery Nulliparas Analysis Univariate logistic regression Total score OR 95% CI P value Multivariate logistic regression Position score OR 95% CI P value Station score OR 95% CI P value SVD IVD CD SVD Multiparas IVD

1.68 1.421.99 .001

0.67 .57.79 .001

0.39 .24 .65 .001

1.59 1.16 2.14 .003

0.64 .47.87 .004

1.78 1.372.31 .001 1.59 1.252.04 .001

0.68 0.53 0.87 .002 0.65 0.51 0.83 .001

0.30 0.16 0.56 .001 0.64 0.28 1.45 .29

2.00 1.273.15 .003 1.30 0.86 1.97 .21

0.48 0.30 0.76 .002 0.81 0.531.23 .31

CD cesarean delivery; CI condence interval; IVD instrumental vaginal delivery; OR odds ratio; SVD spontaneous vaginal delivery.

and station are similar in nulliparas, suggesting that an equal scoring system for these two variables is justied. Median scores at 0, 60, and 90 minutes from the start of the second stage of labor were used to construct a normogram (Figure 1). Only data from women with a score less than 5 at the start of the second stage were used to construct the normogram because a score of 5 at these time points almost universally resulted in spontaneous vaginal delivery (Figure 1). The normogram was used to categorize second stages as normal or abnormal. The ability of progress in the second stage according to the normogram to predict mode of delivery was compared with progress according to time spent in the second stage. For time spent in the second stage, the median value was obtained for nulliparas (67.5 minutes) and multiparas (14 minutes). Any second stage with a duration equal to or less than this median value was classied as normal; any second stage of longer duration was classied as abnormal. Table 3 shows the

ability of normal and abnormal progress as dened by the partogram or by duration of the second stage to predict mode of delivery. The partogram is a better predictor of mode of delivery in both nulliparas and multiparas, although in multiparas the majority of women with an abnormal second stage still have a spontaneous vaginal delivery. Table 4 demonstrates that use of epidural anesthesia, induction of labor, augmentation of labor with oxytocin, dystocia, and prolonged 710 interval are associated with a signicantly reduced probability of a normal second stage of labor, as dened by the normogram that we constructed. Data for fetal outcome are shown in Table 5. Infants born from a normal second stage are lighter than those from an abnormal second stage, although this difference is not statistically signicant in multiparas. Apgar scores do not differ signicantly between infants born from a normal second stage and those born from an abnormal second stage.

Figure 2. Mode of delivery prole according to score at diagnosis of the second stage of labor. SVD spontaneous vaginal delivery; IVD instrumental vaginal delivery; CD emergency cesarean delivery.

VOL. 96, NO. 5, PART 1, NOVEMBER 2000

Sizer et al

Second-Stage Partogram 681

Table 3. Comparison of the Partogram and Time Spent in the Second Stage of Labor as Predictors of Mode of Delivery
Nulliparas Mode of delivery (%) Predictor Progress according to partogram Normal Abnormal Progress according to time in second stage Normal Abnormal SVD 86.2 34.7 IVD 13.8 61.7 CD 0 3.6 Multiparas Mode of delivery (%)

value
2

P value

SVD 98.3 60.2

IVD 1.7 38.3

CD 0 1.5

2 value

P value

205.41

.001

184.91

.001

79.6 39.0

20.2 57.5

0.3 3.5

128.84

.001

99.1 82.3

0.9 17.1

0 0.6

56.32

.001

CD cesarean delivery; IVD instrumental vaginal delivery; SVD spontaneous vaginal delivery.

Discussion
The role of the partogram in the rst stage of labor was established more than 20 years ago, and its practical value as a graphic display of progress, a concise method of conveying information, and a method of recognizing and predicting abnormality through comparison with an ideal prole over time is clearly recognized. It has also been a valuable research tool. A second-stage partogram is a logical extension of the rst-stage partogram and furthers its advantages. Many factors inuence progress in the second stage: the size and shape of the pelvis, which may be related to the height of the mother; birth weight; uterine action; soft-tissue resistance; maternal effort; and degree of exion, caput, and molding of the fetal head.2 However, these factors may all be resolved into the end result of descent and rotation, which allows simplied assessment and the possibility of graphic representation. The validity of this procedure is supported by the positive correlation found between the scores for position and
Table 4. First-Stage Variables and Second-Stage Progress
Nulliparas Second-stage progress (%) First-stage variable Epidural anesthesia Yes No Induction of labor Yes No Augmentation Yes No Dystocia Yes No 710 interval Prolonged Not prolonged Normal 21.5 63.4 36.8 49.2 28.4 54.3 31.8 54.3 23.3 53.0 Abnormal 78.5 36.6 63.2 50.8 71.6 45.7 68.2 45.7 76.7 47.0

station and positive outcome, ie, short duration of the second stage and spontaneous vaginal delivery. Our study also shows that a scoring system based on descent and rotation expressed graphically can chart second-stage progress in the second stage (Figure 1). The distribution of the duration of the second stage reveals that although most of the nulliparas and multiparas in normal labor deliver quickly, a few of these women would benet from a graphic display of their second stage. In women whose labors are complicated, the benet of a partogram should be greater. The style of graphic representation described and the scoring system were readily accepted by the midwifery and medical staff. It was not found to be clinically intrusive, because 88% of women delivered within 1 hour and no further examination was required. Further assessments were needed after 1 hour in only 12% of women whose progress appeared to be slow; the extra clinical information was valuable in the management of these women. A degree of exibility was allowed for in the

Multiparas Second-stage progress (%)

value
2

P value

Normal 38.5 86.7 71.4 80.9 66.2 81.8 62.5 81.7 43.3 81.8

Abnormal 61.5 13.3 28.6 19.1 33.8 18.2 37.5 18.3 56.7 18.2

2 value

P value

122.92

.001

114.76

.001

4.72

.03

2.89

.67

38.07

.001

10.1

.002

31.11

.001

11.91

.001

38.67

.001

26.68

.001

682 Sizer et al

Second-Stage Partogram

Obstetrics & Gynecology

Table 5. Fetal Outcome According to Second-Stage Progress as Determined by the Partogram

Second-stage progress Women Nulliparas Median birth weight (kg) (interquartile range) Apgar score 8 at 5 min (%) Multiparas Median birth weight (kg) (interquartile range) Apgar score 8 at 5 min (%) Normal Abnormal Mann-Whitney U value .001

3.3 (3.0 3.6) 3.5 (3.13.8)

0.6

1.5

.196

3.4 (3.13.8)

3.5 (3.13.8)

.69

must be evaluated in terms of clinical effectiveness, either in a clinical audit or a randomized controlled trial. The partogram must be assessed further for its acceptability to obstetric staff and patients with regard to clarication of recording, identication of abnormality, unnecessary vaginal examinations, and effect on operative delivery rates. The scoring system that may be used to predict problems either from the time of diagnosis of the second stage or after 1 hour must be assessed for its ability to improve outcome for the mother and infant without an unnecessary increase in medical intervention.

0.6

0.8

.797

References
1. Friedman EA. Cervimetry: An objective method for the study of cervical dilatation in labor. Am J Obstet Gynecol 1956;71:1189 93. 2. Friedman EA. Labor: Clinical evaluation and management. 2nd edition. New York: Appleton Century Crofts, 1978. 3. Friedman EA, Sachtleben MR. Station of the presenting part. I. Pattern of descent. Am J Obstet Gynecol 1965;93:5229. 4. Friedman EA, Sachtleben MR. Station of the presenting part. IV. Arrest of descent in nulliparae. Obstet Gynecol 1976;47:129 36. 5. Philpott RH, Castle WM. Cervicographs in the management of labor in primigravidae. 1. The alert line for detecting abnormal labor. J Obstet Gynaecol Br Commonw 1972;79:592 8. 6. Studd JWW. Partograms and normograms of cervical dilatation in the management of primigravid labor. BMJ 1973;4:4515. 7. Kader N, Cruddas M, Campbell S. Estimating the probability of spontaneous delivery conditional on time spent in the second stage. Br J Obstet Gynaecol 1986;93:568 76. 8. Paterson CM, Saunders NStG, Wadsworth J. The characteristics of the second stage of labor in 25,069 singleton deliveries in the North West Thames Health Region. Br J Obstet Gynaecol 1992;99:377 80. 9. Davidson AC, Weaver JB, Davies P, Pearson J. The relation between ease of forceps delivery and speed of cervical dilatation. Br J Obstet Gynaecol 1976;83:279 83.

half-hourly examinations, and omission of nonindicated vaginal examinations did not affect the partogram. The degree of observer agreement was not assessed statistically, but because denition of station and position is a normal midwifery skill, one could argue that a certain degree of observer error must be allowed for in the practical use of the partogram, in the same way that cervical dilatation is accepted in the rst-stage partogram. A scoring system based on position and station appears to differentiate between normal and abnormal labors and therefore satises the criteria under which the rst-stage partogram was introduced.6 However, successful introduction of a second-stage partogram or scoring system depends on how it is to be used. The value of the second-stage partogram is limited. In the 88% of women who deliver by 1 hour, a visual display is unnecessary. Its value is more likely to be appreciated in women whose second stages continue beyond 1 hour and require clarication of the situation. The initial score at the time of diagnosis of the second stage may be used as a predictor of the duration of the second stage and as a predictor of mode of delivery (Figure 2) and may be useful in the early identication of women at increased risk of difcult delivery. A criticism could be made that the time of diagnosis of full dilatation varies among patients for various reasons, thus affecting the time scale. In our study, diagnosis of full dilatation was made by the midwife, as is normal in the labor ward, and the study results should therefore reect progress in the second stage as it occurs in normal practice. The effect of the introduction of second-stage partogram and interventions based on the scoring system

Address reprint requests to:

Andrew Sizer, MB, BCh, PhD Department of Obstetrics and Gynecology University Hospital of Wales Heath Park Cardiff South Glamorgan CF14 4XW United Kingdom E-mail: sizer@cardiff.ac.uk

Received January 24, 2000. Received in revised form May 1, 2000. Accepted May 18, 2000.
Copyright 2000 by The American College of Obstetricians and Gynecologists. Published by Elsevier Science Inc.

VOL. 96, NO. 5, PART 1, NOVEMBER 2000

Sizer et al

Second-Stage Partogram 683