Umbilical cord prolapse – changing patterns and improved outcomes: a retrospective cohort study

Authors

Errata

This article is corrected by:

  1. Errata: Corrigenda Volume 122, Issue 4, 599, Article first published online: 20 February 2015

Abstract

Objective

Umbilical cord prolapse is an acute obstetrical emergency requiring rapid identification and intervention. Its management has undergone significant changes over the past century. This study aims to document the changes in incidence, morbidity, and perinatal mortality over a 69-year period.

Design

A retrospective review of the annual clinical reports of the National Maternity Hospital, Dublin, Ireland, was performed.

Setting

The National Maternity hospital was founded in 1894 and has nearly 10 000 deliveries each year.

Population

All deliveries in the hospital for each year are included in each annual report.

Methods

We reviewed the reports from a 69-year period (1940–2009). Information from the reports was collated into a database and analysed using Microsoft excel 2007.

Main outcome measures

Incidence and outcome of all cases of umbilical cord prolapse were recorded, along with the neurological outcome of all neonatal survivors (available since 1970).

Results

The incidence of cord prolapse has decreased from 6.4/1000 live births in the 1940s to 1.7/100 live births in the last decade. Perinatal survival increased from 46 to 94% in the same period of time. This is inversely related to the use of caesarean section as the recommended method of delivery in this emergency. Short- and long-term neurological impairment remains rare.

Conclusions

There has been a large reduction in the incidence of cord prolapse over a period of 69 years. A reduction in grand multiparity and use of caesarean section as the gold standard for delivery are likely to have accounted for the changes seen. Neurological impairment remains unusual.

Introduction

Umbilical cord prolapse has long been regarded as an acute obstetric emergency, associated with an increased risk of perinatal mortality. The mechanism of fetal demise is through near total or total acute asphyxia,[1] which occurs when the umbilical cord becomes compressed as it lies ahead or alongside the fetal presenting part. The management of cord prolapse has undergone significant changes over the last century, leading to improved maternal and fetal outcomes. As a result of meticulous record keeping in the national maternity hospital, we were able to review 69 years of birth records for changes in incidence, mortality, and perinatal outcomes from this event, and look at the coinciding changes in obstetric management that have brought about these changes. We also reviewed the incidence of long-term neurological disability in infants surviving this event, given the association with intrauterine hypoxia.

Methods

The annual reports of the national maternity hospital were reviewed over a period of 69 years, from 1940 to 2009. The annual reports provide accurate written birth records and also clearly describe births and deaths associated with obstetric emergencies, including umbilical cord prolapse. Detailed perinatal meetings are held every month leading to the publication of an annual clinical report of outcomes. In each year, the method of delivery of each cord prolapse is described, as well fetal outcome in each case. From 1972, detailed records were kept of all cases of neonatal encephalopathy or long-term neurological impairment, and in each case any obvious perinatal insult was described, including cord prolapse, thus allowing the verification of each case of cord prolapse, as these cases would be mentioned in both sections of the report. The data were entered into Microsoft excel and the results were tabulated.

Results

Table 1 demonstrates that there were a total of 1424 cases of umbilical cord prolapse arising from 409 473 viable live births (>28 weeks 1940s to 1980s, >24 weeks 1980 onward) The incidence has decreased decade by decade from 6.4/1000 live births in the 1940s to 1.7/1000 live births in the 2000s.

Table 1. Number and rates of cord prolapse over seven decades, from the 1940s to the 2000s
Decade1940s1950s1960s1970s1980s1990s2000sTotal
Live births19 45534 25850 17174 70577 95068 86384 071409 473
No. cases1252753412281451671431424
Incidence/1000 live births6.48.06.83.11.92.41.73.5

Figure 1 describes patterns of stillbirths and neonatal mortality over the past seven decades. The rate of stillbirths associated with cord prolapse reduced from 48% in the 1940s to 2.1% in the last decade. The neonatal death rate has also reduced, but to a lesser extent, with the rate reducing from 7.2 to 4.2%. Overall survival has improved from 46% in the 1940s to 94% in the 2000s.

Figure 1.

Relative proportion of stillbirths, neonatal deaths, and survival in cases of cord prolapse over seven decades, from the 1940s to the 2000s.

Figure 2 demonstrates the inverse relationship between mortality from cord prolapse and the increasing rate of lower segment caesarean section. Figure 3 plots the changing obstetric management of cord prolapse over time, with lower segment caesarean section becoming the gold standard from the 1960s onwards.

Figure 2.

Cord prolapse, perinatal mortality, and caesarean birth rate over seven decades, from the 1940s to the 2000s.

Figure 3.

Modes of delivery in the case of cord prolapse over seven decades, from the 1940s to the 2000s.

Table 2 demonstrates the rates of neonatal encephalopathy and long-term neurodevelopmental outcomes over the past four decades. There were three children with persistent neurological disability who were described as having choreoathetoid cerebral palsy, ‘severe’ cerebral palsy, and neurodevelopmental delay, respectively, at 1 year of age. Two children with documented encephalopathy in the newborn period were lost to further follow-up.

Table 2. Short- and long-term neurodisability following cord prolapse
Decade1970s1980s1990s2000s
Cord prolapse cases185145167143
Neonatal encephalopathy5513
Persistent neurological disability1200

Discussion

Main findings

We describe the changing pattern in incidence of cord prolapse as well as its perinatal consequences over 69 years in a large tertiary maternity hospital. This large series includes 1424 cases of cord prolapse from almost 440 000 live births. The positive impact of perinatal management on infant survival is clearly illustrated. This reduction in incidence is felt to be linked to the reduction in incidence of grand multiparity, which is a significant risk factor for cord prolapse. Seventy-five percent of cord prolapse events in the 1940s occurred in multiparous women, and this has remained a significant risk factor today, with 65% of events occurring in this group, although our overall incidence of grand multiparity (>4) is now <1%.

The major contributing factor to neonatal survival has been the reduction in intrapartum stillbirth, which has seen a 20-fold improvement over the period reviewed. This reduction is clearly related to the introduction of lower segment caesarean section as the standard of care in the management of this obstetric emergency, with a corresponding decrease in all other methods of delivery. Early vaginal examination of women upon presentation in labour is also felt to have improved the detection of cord prolapse.

This study also addressed the issue of neurological outcome in infants following delivery. Out of 683 cases of prolapse since 1972 (when neurological outcome was recorded), there have been 14 cases of neonatal encephalopathy and three cases of documented cerebral palsy. The incidence of neonatal encephalopathy is 2%, which is similar to previous reports,[2, 3] with an incidence of cerebral palsy of 0.43%.

Strengths and limitations

The major strength of this study lies in the size and time span of the cohort, covering seven decades of a single tertiary centre. This allowed us a truly fascinating insight into the changes in obstetric care over time and lets us conclusively answer the important question of the impact of this event on the newborn.

Its limitations lie in the fact that this study was retrospective and thus we can only report on the information contained in the annual reports. Unfortunately, we are unable to comment on the gestation age of the cohort as this was only documented in the case of a perinatal death. We have noted that 83% of perinatal deaths associated with cord prolapse in the last decade were <28 weeks of gestation at delivery, with documented causes of death being related to complications from prematurity rather than as a direct result of the prolapse. We also cannot comment on how many births were at full dilatation, as this was not recorded in the reports. Information collected varied over time, and neurological outcomes were only available from the 1970s, with currently used terminology only applied in the last 20 years of the records. This makes the long-term outcomes of these infants very difficult to comment on.

Interpretation

When the pathophysiology of cord prolapse is reviewed it would seem that it is akin to an ‘all or nothing event’, either causing overwhelming neurological injury and death, or causing little or no cerebral injury. Clinical, animal, and neuroimaging studies indicate that different types of insult lead to various patterns of brain injury and neurological outcome.[4-6] This is explained by the autoregulatory changes that occur during asphyxia, as well as the metabolic requirements of different regions of the brain.[7]

In the case of longer lasting insults causing subacute asphyxia, the natural response of the fetus is to prioritise cerebral blood flow and reroute blood flow from areas of low metabolic demand to areas of higher metabolic demand.[8] The brainstem is relatively spared, with early loss of white matter and the cerebral cortex. This sequence of events results in the most common pattern of neonatal brain injury, with the deep grey nuclei and the parasaggital white matter most affected on cranial ultrasound and early magnetic resonance imaging (MRI).[7] It also fits with the clinical pattern of spastic cerebral palsy, which can affect these infants. In contrast, the acute near total asphyxia caused by cord prolapse causes a failure of normal autoregulation as a result of bradycardia and hypotension. The acute nature of the insult also quickly leads to the failure of cerebral blood redistribution, with cell death in the most metabolically active area of the brain – the brainstem.[9] This is frequently incompatible with life. Thus, it can be concluded that if the brainstem has not been affected then it is likely that the less active areas of the white matter and deep grey nuclei will also be preserved, and that the infant will not have an adverse neurological outcome.

We did notice a slight reduction in the number of cord prolapse cases being delivered by caesarean section over the last decade. Once again we are limited by the lack of data regarding gestation and stage of labour in interpreting this finding. Those delivered by vaginal delivery, whether operative, breech, or normal, regardless of gestation, had a 90% survival rate (27/30). The three deaths were attributable to prematurity or congenital abnormality. Thus delivery by methods other than caesarean section is a safe and well-established means of delivery in our centre.

There is currently no data from prospective or randomised controlled trials regarding the management of cord prolapse because of the infrequent and urgent nature of this condition. There are multiple guidelines available on its management,[10, 11] with the majority of recommendations graded B–D given the lack of high-quality data. The focus of these guidelines is the early identification of patients at risk, the early diagnosis of the condition, the relief of pressure from the cord itself through positioning and direct removal of the pressure of the presenting part, and an expedited delivery. The only area where our management differs from the main guidelines is regarding the use of tocolytics, which we do not routinely use.

Conclusion

This analysis demonstrates that the incidence and consequences of cord prolapse has been influenced by both maternal demographics, in relation to multiparity, and by management, related to ready access to caesarean section and neonatal resuscitation. Women with unstable lies, the commonest predisposing cause, are routinely admitted to hospital at 37–38 weeks of gestation to await stabilisation of the fetal position. In addition, high parity associated with the occurrence of unstable lie has fallen from >20% in the 1940s to 0.7% in 2011. The finding of a low neurological morbidity among survivors provides insight into the pathogenesis of severe asphyxia and neonatal encephalopathy.

Disclosure of interests

The authors have no competing interests or relevant disclosures to make.

Contribution to authorship

CG performed data collection, analysed data, and wrote the article. C0'H contributed to project planning, reviewed the data, and edited the article. JFM conceived the research, identified the data source, reviewed the data, and aided in editing and writing the article.

Details of ethics approval

Ethics approval was waivered as this study was undertaken using published population-based data in the form of the annual reports of the hospital. No specific patient data were accessed for this review.

Funding

There was no funding provided for this study.

Acknowledgements

We would like to thank the librarian of the national maternity hospital for her help in accessing the annual reports. We would also like to thank Dr Michael Robson for his support of this project during his tenure as master of the hospital.

Ancillary