Khalid S. Khan  ( Department of Obstetrics and Gynaecology, Aga Khan University Medical Centre, Karachi. )
Asad Rizvi  ( Department of Obstetrics and Gynaecology, Aga Khan University Medical Centre, Karachi. )
Javaid H. Rizvi  ( Department of Obstetrics and Gynaecology, Aga Khan University Medical Centre, Karachi. )

To determine if prolonged active phase of labour is associated with increased risk of uterine scar rupture in labour following previous lower segment caesarean section,a retrospective cohort study (1988-91) was done to analyse active phase partographs of 236 patients undergoing trial of labour following caesarean section, 7 (3%) of whom had scar rupture. After onset of active phase (3 cm cervical dilatation), a 1 cm/h line was used to indicate “alert”. A zonal partogram was developed by dividing the active phase partographs into 5 time zones: A (area to the left of “alert” line), B (0-1 h after “alert” line), C(1-2 h after “alert” line), D (2-3 h after “alert” line) and EF (>3 h after “alert” line), The relative risk of uterine scar rupture was calculated for different partographic time zones. The relative risk of uterine scar rupture was 10.5 (95% confidence interval 1.3-85.5, p=O.Ol) at 1 hour after crossing the “alert” line; 8.0(95% confidence interval 1.6-40.3, p=0.009) at 2 hours after crossing the “alert” line; and 7.0(95% confidence interval 1.6-29, p=O.O2) at 3 hours after crossing the “alert” line. In women undergoing trial of labour following caesarean section, prolonged active phase of labour is associated with increased risk of uterine rupture. A zonal partogram may be helpful in assessing this risk in actively labouring women who cross the partographic “alert” line (JPMA 46:120, 1996).

In the prevention of uterine rupture in women undergo­ing a trial of labour following caesarean section, timely diagnosis and prompt management of labour dystocias has been recommended¹.. Chazotte and Cohen² have commented that “arrest disorders might indicate or predispose a patient to uterine rupture, a trial of labourshouldbe discontinued if there is no prompt response to uterine stimulation”. However, there is controversy about the upper limit of duration of normal active phase of labour³ and the length of labour beyond which the increased risk of uterine rupture in women with previous caesarean section warrants intervention is not clearly known.
This retrospective cohort study evaluates the relation­ship between prolonged labour during the active phase and uterine scar rupture in women having a trial of labour following a previous caesarean section. “prolonged labour” is defined as crossing of a partographic “alert” line during active phase and “uterine rupture” is defined as a uterine scar separation necessitating operative intervention lateral pelvimetry. Women with more than one caesarean, a classical caesarean, placentapreviaorrefusal to undergo a trial had a repeat elective caesarean (n=152) and were excluded. Partographic records could not be retrieved in 10 cases, 9 of vaginal deliveiy and 1 of caesarean section due to “fetal distress”. Thus, in total 236 ase.s were evaluated.

The AgaKhan University Department of Obstetrics and Gynaecology provides secondaiy and tertiary level care to a largely self- referred population. During 1988-91, out of 398 consecutive patients with previous caesarean section, 246 (67.8%) were selected for a labour trial based on the presence ofa non- recurrent cause forcaesareansection and an adequate maternal pelvis, as judged by clinical assessment and erect
All patients signed an informed-consent and were managed according to a uniform policy during labour. Partograms were commenced according to the WHO guide­ lines⁴ . Vaginal examination were performed every 2-4 hours or earlier if indicated. After the diagnosis of active phase (labour established with regular uterine contractions and cervical dilatation of 3 cm) a 1 cm/hour line was used as the “alert” line. Fetal heart rate was continuously monitored throughout labour. Amniotomy and oxytocin were used for the same indications as for patients with unscarred uteri. The initial dose of oxytocin, 1 mIU/min, was doubled eveiy 30 min until the patient had adequate uterine activity (3-4 contractions in 10 min, each lasting for 45 seconds) or until the maximum dose of32 mIU/min was reached.
Partograms were analysed by drawing at each hour after the “alerf line 3 lines which divided the partogram into zones A,B,C,D,E and F as shown in Figure.

In the definition of prolonged active phase there is controversy about the lag time after the “alert” line³. We, therefore, used three different lag times.afterthe “alert” line in analysis: firstwith 1 hourlagtime making zones A-B normal and C-F prolonged; second with 2 hour lag time making zones A-C normal and D-F prolonged; and third with 3 hour lag time making zones A-D normal and E-F prolonged. Our null hypothesis was one of no association between active phase labour prolongation and uterine scar rupture.
We tested the null hypothesis for each one of the three lag times by calculating and comparing rupture rates for normal and prolonged labour. The differences in baseline characteristics were tested with analysis ofvariance andtestof linear trend. Relative risk (95% confidence inteivals) and Fisher’s exact test were used for analysing the differences in rate of scar rupture according to different definitions of prolonged active phase. Statistical analysis was performed with Epi Info software (version 6.0, Centres for Disease ContrOl, Atlanta, U.S.A.).

Successful vaginal delivery was achieved in 190 (77.2%) out of the total of 246 cases. Of the rest (56), 26 cases had repeat caesarean section for slow progress of labour in the first stage, 2 had caesarean sections for non-descent of foetal head in the second stage, 27 had caesarean sections for “fetal distress” (abnormal cardiotocographic trace with or without meconium and fetal scalp pH of <7.2) and 1 had caesarean section for suspected scar rupture during the second stage of labour. There were-7 (3%) cases of uterine scar rupture.
The general characteristics of the 236 subjects whose partograms were analysed are shown in Table I.

There was no difference in the distribution of the age, parity, gestational age at delivery and birth weight in the different partographic zones withthe exceptionof use ofoxytocin. Outofthe 7 cases of scar rupture, 1 patient was in zone B (1-2 hour period beyond the “alert” line), 3 patients were in zoneD (2-3 hourperiodbeyond the “alert” line) and 1 patient each was in zones E & F (beyond the3 hourline).


Table II shows the relative risks (95% confidence intervals) of having a uterine scar rupture if the labour was prolonged according to different lag times afterthe “alert” line.
The risk increased by 10.5 times (95% confidence interval 1.3-85.5) for the 1 hour lag time after “alert” line; by 8 times (95%confidence interval 1.6-40.3) for the 2 hour lag time after “alert” line; and by 7 times (95% confidence interval 1.6-29) for the 3 hour lag time after “alert” line.

The aetiology of uterine scar rupture during trial of labour following caesarean section is believed to arise from excessive tension generated during the active phase, which has a cumulative effect as labour becomes prolonged. How­ever, until recently this association was not subjected to epidemiological scrutiny. Leung et al⁵ were probably the first to show in a case-control study, an association between prolonged labour and uterine rupture in cases of trial of labour following previous caesarean section (women with uterine scar rupture were more likely to have been exposed to labour dysfunction than the control group with odds ratio: 7.2,95% confidence interval: 2.6-19.8). In their paper, prolonged labour was defined as active phase protraction (cervical dilatation rate <1.5 cm/h) or arrest of dilatation for >2 h. However, the measurement of duration of labour or the rate of its progress is subjective, particularly the demarcation be­tween the end of latent phase and the onset of active phase can have large inter/intraobserver variation. The criteria for defining onset of active phase were not cleariy delineated by Leung et al⁶ raising doubts about the accuracy of assessing exposure to prolonged labour.
In our cohort study, the relative risk of uterine rupture in cases of prolonged labour was similar in size to the odds ratio reported by Leung et al⁵ but the confidence intervals in our analysis were larger. This association between prolonged labour and uterine scar rupture cannot be explained on the basis of chance alone as the p value was <0.05 and the lower confidence limit was above one, but there are possible confounders and biases. In our analysis, the confounders were equally distributed mall the partographic zones except the use of oxytocin which showed a linear trend with prolonging labour (Table I). This is not unusual as the use of oxytocin to augment slow labour is considered safe in patients with previous caesarean section. We also attempted to minimise the biases arising from the difficulty in accurate assessment of exposure to prolonged active phase by the use of a more objective tool for assessing progress of labour, the partognun.
The precision of our findings is limited by the small sample size but it provides a useful frame-work to analyse the active phase partograms in trial oflabourfollowing caesarean section. The next step should be to cariy out further prospec­tive cohort studies with objective and unbiased assessment of labour progress in order to determine the length of labour beyond which the risk of uterine rupture (and its complica­tions) becomes far too high to exceed the expected benefits of vaginal delivery. We would, therefore, like to invite other researchers in Pakistan to perform similar analyses of parto­grams in women undergoing trial of labour following caesar­ean section in their obstetric units. Such a collaborative effort would be the first step in the right direction in developing nation-wide guidelines for the safe conduct of labour in women with previous caesarean section.
The aetiology of uterine scar rupture during trial of labour following caesarean section is believed to arise from excessive tension generated during the active phase, which has a cumulative effect as labour becomes prolonged. How­ever, until recently this association was not subjected to epidemiological scrutiny. Leung et al5 were probably the first to show in a case-control study, an association between prolonged labour and uterine rupture in cases of trial of labour following previous caesarean section (women with uterine scar rupture were more likely to have been exposed to labour dysfunction than the control group with odds ratio: 7.2,95% confidence interval: 2.6-19.8). In their paper, prolonged labour was defined as active phase protraction (cervical dilatation rate <1.5 cm/h) or arrest of dilatation for >2 h. However, the measurement of duration of labour or the rate of its progress is subjective, particularly the demarcation be­tween the end of latent phase and the onset of active phase can have large inter/intraobserver variation. The criteria for defining onset of active phase were not cleariy delineated by Leung et al6 raising doubts about the accuracy of assessing exposure to prolonged labour.
In our cohort study, the relative risk of uterine rupture in cases of prolonged labour was similar in size to the odds ratio reported by Leung et al5 but the confidence intervals in our analysis were larger. This association between prolonged labour and uterine scar rupture cannot be explained on the basis of chance alone as the p value was <0.05 and the lower confidence limit was above one, but there are possible confounders and biases. In our analysis, the confounders were equally distributed mall the partographic zones except the use of oxytocin which showed a linear trend with prolonging labour (Table I). This is not unusual as the use of oxytocin to augment slow labour is considered safe in patients with previous caesarean section. We also attempted to minimise the biases arising from the difficulty in accurate assessment of exposure to prolonged active phase by the use of a more objective tool for assessing progress of labour, the partognun.
The precision of our findings is limited by the small sample size but it provides a useful frame-work to analyse the active phase partograms in trial oflabourfollowing caesarean section. The next step should be to cariy out further prospec­tive cohort studies with objective and unbiased assessment of labour progress in order to determine the length of labour beyond which the risk of uterine rupture (and its complica­tions) becomes far too high to exceed the expected benefits of vaginal delivery. We would, therefore, like to invite other researchers in Pakistan to perform similar analyses of parto­grams in women undergoing trial of labour following caesar­ean section in their obstetric units. Such a collaborative effort would be the first step in the right direction in developing nation-wide guidelines for the safe conduct of labour in women with previous caesarean section.

The authors am grateful to Professor S.C. Robinson for his encouragement and support of this project and to all the consultants and staff for their untiring efforts in the manage­ment of patients with trial of labour following caesarean section during the course of this study.

1. Farmei R.M., Kirschbaum. 1., Potter. D. d al. Uterine rupture during that of labour after previous caesar ean section. Am. J. Obstet. Gynecol., 1991;165:996. 1001.
2. Chazotte, C. and Cohen, W.R. Catastrophic complications ofprevious caesarean section. Am. J. Obstd. Oynecol., 1990;163:738-742.
3. Crowthe, C.,Enkin,M..Keirse,M.N.J.C.Monitoring the progress of labour. In: Effective Care in Pregnancy and Childbirth Vol. 2 (Chalmers, I.. Kierse, M.N.J.C. and Enkin. M,). Oxford, Oxford Univ. Press, 1989, pp. 833.845.
4. World Health Organisation. The partograph: A managerial tool for the prevention of prolonged labour. Section II: A Users Manual. Geneva, WHO, 1988.
5. Leung. AS., Farmer, R.M., Leung, E.K. et al. Risk factors associated with uterine rupture during trial of labour alter caesarean delivezy. A case.control study. Am.J. Obstet. Gynecol., 1993;168:1358-1363.