Accepted Manuscript

Dysfunctional Uterine Activity in Labour and Premature Adverse Cardiac Events:

population-based cohort study

Alyssa Kahane, BSc, Alison L. Park, MSc, Joel G. Ray, MD MSc

PII: S0828-282X(17)31081-4
DOI: 10.1016/j.cjca.2017.10.007
Reference: CJCA 2623

To appear in: Canadian Journal of Cardiology

Received Date: 25 August 2017

Revised Date: 4 October 2017
Accepted Date: 11 October 2017

Please cite this article as: Kahane A, Park AL, Ray JG, Dysfunctional Uterine Activity in Labour and
Premature Adverse Cardiac Events: population-based cohort study, Canadian Journal of Cardiology
(2017), doi: 10.1016/j.cjca.2017.10.007.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
 ACCEPTED MANUSCRIPT

Dysfunctional Uterine Activity in Labour and

Premature Adverse Cardiac Events: population-based cohort study

Alyssa Kahane, BSc

PT
St. Michael’s Hospital, University of Toronto, Toronto, Ontario
Faculty of Medicine, University of Ottawa, Ottawa, Ontario

RI
Alison L Park, MSc
Institute for Clinical Evaluative Science

SC
Toronto, Ontario

Joel G Ray, MD MSc

U
Departments of Medicine,
Health Policy Management and Evaluation,
AN
and Obstetrics and Gynecology,
St. Michael’s Hospital, University of Toronto, Toronto, Ontario
M
D
TE
EP

Contact:
Joel G Ray
Department of Medicine, St. Michael’s Hospital
C

30 Bond Street
Toronto, Ontario
AC

M5B 1W8
Tel: (416) 864-6060, Ext 77442
e-mail: rayj@smh.ca

Abstract count: 263

Text word count: 2173
 ACCEPTED MANUSCRIPT

Brief summary
We assessed the risk of heart failure, cardiomyopathy or dysrhythmia in women with a history
of prolonged labour. We conducted a population-based cohort study in the entire province of
Ontario, including all singleton livebirths at 24 to 41 weeks’ gestation, between 1992 and 2016.

PT
Women with prolonged labour dif not appear to be at a higher future risk of the heart failure,
cardiomyopathy or dysrhythmia.

RI
U SC
AN
M
D
TE
C EP
AC
 ACCEPTED MANUSCRIPT

Abstract
Background: There exist structural and physiological commonalities between myometrial and
myocardial tissue, and each can become dysfunctional, such as in the presence of
cardiometabolic factors.

PT
Methods: This population-based cohort study comprised 1,608,720 women with ≥ 1 singleton
hospital livebirths at 24-41 weeks’ gestation in Ontario, 1992-2016. The main exposure was
prolonged first stage of labour; secondary exposure was prolonged second stage of labour. The

RI
main outcome was a composite of heart failure, cardiomyopathy or dysrhythmia ≥ 1 day after

SC
the index delivery. Cox proportional hazard regression was used to generate a hazard ratio
(HR), adjusted for maternal age, parity, obstructed labour or fetal malposition, preeclampsia,
income quintile, rural residence, preterm birth and infant birthweight -- each at the time of

U
delivery; time-varying drug/tobacco use, obesity, diabetes mellitus, chronic hypertension,
AN
kidney disease, dyslipidemia -- each diagnosed before or at time of delivery; as well as newly
diagnosed coronary artery disease or congenital heart disease arising ≥ 1 day after the index
M

delivery.
Results: After a median follow-up of 10.5 and 14.0 years, respectively, there were 78 composite
D

cardiac events (2.33 per 10,000 person-years) among women with prolonged first stage of
labour vs. 4114 events (2.30 per 10,000 person-years) among those without prolonged labour –
TE

a crude HR of 1.07 (95% CI 0.86 to 1.34) and an adjusted HR (aHR) of 1.09 (95% CI 0.87 to 1.36).
Women with prolonged second stage of labour had an aHR of 0.86 (95% CI 0.75 to 0.99) for the
EP

composite outcome.
Conclusions: Women with prolonged labour do not appear to be at a higher short-term risk of
C

cardiac outcomes.
AC

Keywords: Prolonged labour; uterine inertia; heart failure; cardiac dysrhythmia;

cardiomyopathy.
 ACCEPTED MANUSCRIPT

Introduction

In pregnancy, uterine adaptation occurs with advancing gestational age, in preparation for eventual

labour, including an increase in myometrial gap junction proteins, to strengthen cell to cell coupling

and coordinated contractility between neighbouring cells <1>. Interestingly, the functional syncytia of

PT
the labouring myometrium and human myocardium bear some similarities <2> (Supplementary file 1).

Others have even proposed the autologous implantation of non-human uterine myometrium into

RI
large areas of infarcted myocardium <3>. Unlike the functionally specific sinoatrial node of the heart,

SC
any myometrial cell can act as a pacemaker <4>, with depolarization of neighbouring cells and

coordinated regional contraction, followed by whole-organ contraction and eventual cervical dilation

with advancing labour <5>.

U
AN
In the heart, mutations in genes encoding calcium ion channels have been identified in individuals
M

with certain cardiac dysrhythmias and cardiomyopathies <6, 7>. Mutations in myocardial potassium

channels have also been associated with dysrhythmias <8>. While mutations in potassium channels
D

would theoretically be expected to alter myometrial contractility, clinical data related to abnormal
TE

uterine function during childbirth are lacking <9>.

Cardiometabolic (metabolic syndrome) factors are thought to heighten the risk of heart failure <10>
EP

and cardiac dysrhythmias <11>. In addition to advanced maternal age, some cardiometabolic features
C

– obesity and the hypertensive disorders of pregnancy – are associated with prolonged first stage of
AC

labour <12-14>. A prolonged second stage of labour is more common in women with gestational

diabetes mellitus, advanced maternal age and fetal macrosomia <15, 16>.

Given that there are some structural and physiological commonalities between myometrial and

myocardial tissue, and that each can become dysfunctional in the presence of cardiometabolic

factors, we evaluated whether women who had prolonged labour are at a higher short-term future
 ACCEPTED MANUSCRIPT

risk of heart failure, cardiomyopathy or dysrhythmias, compared to those who experienced normal

labour. The reasons for testing this hypothesis were twofold. First, the concept is novel, biologically

plausible, and not previously assessed in any epidemiological manner. Second, if an association is

seen, then this might introduce a prolonged labor as a risk factor for heart disease in women of

PT
childbearing age, especially in the early years after an affected birth.

RI
Methods

SC
Study Design

We conducted a retrospective population-based cohort study using healthcare administrative data

U
for the province of Ontario. Datasets were linked using unique encoded identifiers and analyzed, at
AN
the Institute for Clinical Evaluative Sciences (ICES). All participants were enrolled in the Ontario Health
M

Insurance Plan (OHIP), which covers all aspects of antenatal care, including maternal serum screening

and ultrasonography, as well as hospital and postnatal care. This study was approved by the
D

institutional review board at Sunnybrook Health Sciences Centre, Toronto, Canada.

TE

Participants
EP

Included were all Ontario women aged 16 to 49 years, with an obstetrical delivery of a singleton
C

livebirth at ≥ 24 weeks’ gestation, between April 1, 1992 and March 31, 2015. The end follow-up date
AC

was March 31, 2016, allowing for a maximum follow-up of 24 years. Excluded were women diagnosed

with any cardiac disease ≤ 24 months prior to, and including, the index birth hospitalization. Also

excluded were non-Ontario residents, those without a valid OHIP health card number, missing rurality

status or discharge date at the time of the index delivery, and those whose last date of contact or

5
 ACCEPTED MANUSCRIPT

death occurred before the index delivery discharge date (Supplementary file 2). The unit of analysis

was the woman, while allowing for more than one delivery per woman during the study period.

Exposures and outcomes

PT
The main exposure of interest was a diagnosis of any prolonged first stage of labour during the index

birth hospitalization (Supplementary file 2). Secondary exposures were i) a prolonged second stage

RI
labour (secondary exposure #2), ii) any prolonged labour and/or uterine inertia (secondary exposure

SC
#3), and iii) any uterine inertia (secondary exposure #4). A description of the prolonged stages of

labour is shown in Supplementary file 3. A prolonged first stage of labour was chosen as the main

U
exposure because it more likely to reflect uterine inertia (the absence of effective uterine
AN
contractions during labour); whereas, a prolonged second stage can be attributed not only to uterine
M

inertia, but also to fetal malposition or cephalopelvic disproportion, for example.

During the study period, if a woman had more than one delivery, then her exposure status was
D

updated at each subsequent delivery, as a time-varying non-reversible exposure in each model. The
TE

reference group was women who remained unexposed to any prolonged labour or uterine inertia.

The primary composite outcome was a hospitalization for newly diagnosed heart failure,
EP

cardiomyopathy or cardiac dysrhythmia, arising ≥ 1 day after the index birth discharge date (“time
C

zero”) (Supplementary file 2).

AC

All study exposures and outcomes were identified using the International Classification of Diseases

(ICD) coding system (ICD-9 before 2002 and ICD-10-CA thereafter).

Database sources

6
 ACCEPTED MANUSCRIPT

All hospitalizations and procedures, including obstetrical deliveries and cardiac outcome events,

were identified using the Canadian Institute for Health Information’s Discharge Abstract Database

(DAD). The DAD contains the unique encrypted healthcare number, age and sex of the participant,

date of admission, and up to 16 diagnoses coded by ICD-9, and up to 25 diagnosis coded by ICD-10-

PT
CA. Outpatient variables were identified in the OHIP Database (Supplementary file 2). The OHIP

database contains records of all physicians’ billing information for outpatient and inpatient services,

RI
including a service date and a single diagnosis. Both databases were also used to identify exclusion

SC
criterion of pre-existing cardiac disease or dysrhythmias ≤ 24 months prior to, and including, the first

index birth hospitalization, as well as covariates (Supplementary file 2).

U
Maternal mortality was identified from the Ontario Ministry of Health and Long Term Care’s
AN
Registered Persons Database, which contains demographic information and encrypted healthcare
M

numbers for all individuals eligible for OHIP. Dissemination area income quintile and rurality were

based on Statistics Canada census data.

D
TE

Statistical analysis

Time-to-event analyses were conducted using multivariable Cox regression models with time-
EP

varying covariates, to derive a hazard ratio (HR) and 95% confidence interval (CI) for each study
C

outcome, comparing women with a prolonged labour to those without any prolonged labour or
AC

uterine inertia of any type (the referent). Censoring was on death or arrival at the end of study period

of March 31, 2016. The HRs were adjusted for maternal age (continuous in years), parity (0, 1, ≥ 2),

obstructed labour or fetal malposition, preeclampsia or eclampsia, residential income quintile (1, 2, 3,

4, 5, unknown), rural residence, current preterm birth and infant birthweight (continuous in grams) --

each at the time of a given delivery; and time-varying drug or tobacco use, obesity, diabetes mellitus,

7
 ACCEPTED MANUSCRIPT

chronic hypertension, kidney disease, dyslipidemia -- each diagnosed before or at the time of a given

delivery; as well as newly diagnosed coronary artery disease or congenital heart disease arising at

least 1 day after the index delivery discharge date (i.e., time zero). The proportional hazards

assumption was assessed by a Wald test for interaction between the exposure and a function of

PT
survival time, which did not detect a significant departure.

The main model was further stratified by parity (nulliparous vs. parous), prior Caesarean delivery

RI
(limited to parous women), spontaneous vaginal delivery requiring minimal or no assistance, induced

SC
labour, and forceps or vacuum assisted vaginal delivery (additional analysis 1).

While time zero in the main model was defined as starting 1 day after the index birth discharge date,

U
two additional analyses were done evaluating the composite outcome with time zero starting 1) at 91
AN
days after the index birth discharge date (additional analysis 2), and 2) at the index delivery
M

hospitalization date (additional analysis 3). Finally, additional analysis 4 comprised the same approach

as in the main model, but the HRs were no longer adjusted for newly diagnosed coronary artery
D

disease or congenital heart after the index delivery discharge date.

TE

All statistical analyses were performed using SAS version 9.4 for UNIX (SAS Institute Inc.).
EP

Results
C

Of 1,744,055 women initially eligible, 135,335 (7.8%) were excluded. Thus, the final cohort
AC

comprised 1,608,720 women (Table 1). The reasons for study exclusion are listed in Supplementary

File 4. At the first delivery during the study, the mean maternal age at delivery of the women in this

cohort was 28.6 years. The demographic characteristics of the 23,318 women exposed to prolonged

first stage of labour were generally similar to the 1,345,054 women. Of all women with a prolonged

first stage of labour, 21,732 (90.6%) were nulliparous, in contrast to 1,117,330 (79.6%) women

8
 ACCEPTED MANUSCRIPT

without any prolonged labour or uterine inertia. In terms of survival during this study, 125 women

(0.5%) in the exposed cohort and 11,916 women (0.9%) in the unexposed cohort died during the

follow-up. The median duration of follow-up was 10.5 years among women with a prolonged first

stage of labour; 12.8 years among those with a prolonged second or unspecified stage of labour, or

PT
uterine inertia; and 14.0 years among unexposed women (Table 1).

The composite cardiac outcome occurred in 78 women with prolonged first stage of labour (2.33 per

RI
10,000 person-years) and in 4114 women (2.30 per 10,000 person-years) with non-prolonged labour,

SC
yielding a crude HR of 1.07 (95% CI 0.86 to 1.34) and an aHR of 1.09 (95% CI 0.87 to 1.36) (Table 2).

Among those with a prolonged first stage of labour, 46 outcome events were for newly diagnosed

U
heart failure or cardiomyopathy, and 33 events for a cardiac dysrhythmia. Upon stratifying the main
AN
model by pregnancy and delivery factors, the results remained non-significant (additional analysis 1;
M

Figure 1). Evaluating the composite outcome starting at 91 days after the index birth discharge date

(additional analysis 2; Supplementary File 5) and at the index delivery hospitalization (additional
D

analysis 3; Supplementary File 6) also did not change the main findings. Removing newly diagnosed
TE

coronary artery disease or congenital heart after the index delivery from the main model did not

appreciably alter the adjusted HRs (additional analysis 4; Supplementary File 7).
EP

Further analysis of the composite cardiac outcome by type of prolonged labour showed similar
C

findings to the main model (Table 3). The exception was a slightly lower risk of the composite
AC

outcome in relation to a prolonged second stage of labour (aHR 0.86, 95% CI 0.75 to 0.99) or any

prolonged labour and/or uterine inertia (aHR 0.92, 95% CI 0.85 to 0.89) (Table 3).

9
 ACCEPTED MANUSCRIPT

Discussion

Among a large cohort of women cared for within a universal health system, no short-term increased

risk of heart failure, cardiomyopathy or cardiac dysrhythmia was seen within the first decade after a

pregnancy affected by a prolonged first stage of labour.

PT
This is the first population-based cohort study to examine the relation between prolonged labour

and cardiac outcomes. The study included all livebirths at ≥ 24 weeks’ gestation, producing a large

RI
and diverse cohort of women cared for within a universal healthcare system. We accounted for

SC
several factors known to influence the risk of prolonged labour, but others were not accounted for,

such as maternal body mass index or intrapartum fetal heart rate. The diagnostic codes used herein to

U
define prolonged labour have not been previously validated. Some women herein may have been
AN
misclassified as having experienced a prolonged first or second stage of labour <17, 18>, thereby
M

heightening the likelihood of a null effect, providing that their exposure misclassification was non-

differential. Although rare, it is also possible that some women who developed cardiac disease and
D

died out of hospital were missed in the exposed and unexposed groups. Conversely, those diagnosed
TE

with heart failure, a cardiomyopathy or a cardiac dysrhythmia in an outpatient setting were not

considered herein, but they would certainly reflect a more mild form of any of these cardiac
EP

conditions. For example, women with peripartum cardiomyopathy tend to be ill, requiring
C

hospitalization <19>, and few die within the first 40 months <20>.
AC

We saw no association between a prolonged first stage of labour and short-term cardiac outcomes

within 10 to 13 years after delivery. We must acknowledge that a typical woman in this study, at

maximum follow-up, would still be pre-menopausal, before heart disease is likely to manifest,

especially that related to cardiometabolic risk factors. Additionally, among women with a prolonged

first stage of labour, the prevalence of common cardiometabolic risk factors within 24 months before
 ACCEPTED MANUSCRIPT

index the delivery date was only 3.4% for obesity, 5.9% for diabetes mellitus, and 3.8% for chronic

hypertension. The 2% rate of tobacco use observed herein is much less than the 10% rate reported

among surveyed pregnant women in Ontario <21>. While the women in this study reflect Ontarian

women who conceive and maintain a pregnancy, those who do not conceive may have a different

PT
cardiometabolic or health profile.

Herein, there was a borderline lower risk in women with a prolonged second stage. While there are

RI
clear differences between the first and second stage of labour, as well as a few commonalities

SC
between the myometrium and myocardium (Supplementary File 1; Supplementary File 5), we cannot

explain why there would be any difference in cardiac risk by the stage at which labour is prolonged, if

real.
U
AN
In conclusion, the current study findings do not suggest that a history of prolonged labour heightens
M

the short-term risk of myocardial failure or electrical disturbance.

D

Disclosure of interests: The authors report no conflict of interest concerning the materials or
TE

methods used in this study or the findings specified in this paper.

EP

Contribution to Authorship: JR was involved in the conceptualization of the study, design and
C

methodology, protocol review and editing, writing and editing of the manuscript. AK was involved in
AC

the design and methodology, writing and editing of the manuscript. AP was involved in the study,

design and methodology, protocol review and editing, statistical analysis, writing and editing of the

manuscript.

11
 ACCEPTED MANUSCRIPT

Acknowledgements: This study was supported by the Institute for Clinical Evaluative Sciences (ICES),

which is funded by an annual grant from the Ontario Ministry of Health and Long-Term Care

(MOHLTC). The opinions, results and conclusions reported in this paper are those of the authors and

are independent from the funding sources. No endorsement by ICES or the Ontario MOHLTC is

PT
intended or should be inferred.

RI
U SC
AN
M
D
TE
C EP
AC

12
 ACCEPTED MANUSCRIPT

References
1. Riemer RK, Heymann MA. Regulation of Uterine Smooth Muscle Function during Gestation. Pediatr
Res 1998;44:615-627.
2. Andersen HF. Why the heart is like an orchestra and the uterus is like a soccer crowd. Am J Obstet
Gynecol 2016;214:300.

PT
3. Taheri S, Yeh J, Batt R, et al. Uterine myometrium as a cell patch as an alternative graft for
transplantation to infarcted cardiac myocardium: a preliminary study. Int J Artif Organs 2008;31:62-

RI
67.
4. Banney D, Young R, Paul J, et al. A hypothesis for self-organization and symmetry reduction in the

SC
synchronization of organ-level contractions in the human uterus during labor. Symmetry 2015;7:1981-
1988.

U
5. Priori SG, Wilde AA, Horie M, et al. HRS/EHRA/APHRS Expert Consensus Statement on the Diagnosis
and Management of Patients with Inherited Primary Arrhythmia Syndromes. Journal of Arrhythmia
AN
2014;30:1-28.
6. Napolitano C, Antzelevitch C. Phenotypical Manifestations of Mutations in the Genes Encoding
M

Subunits of the Cardiac Voltage-Dependent L-Type Calcium Channel. Circ Res 2011;108:607-618.
7. Kjenseth A, Fykerud TA, Sirnes S, et al. The Gap Junction Channel Protein Connexin 43 Is Covalently
D

Modified and Regulated by SUMOylation. J Biol Chem 2012;287:15851-15861.

TE

8. Greenwood IA, Tribe RM. Kv7 and Kv11 channels in myometrial regulation. Exp Physiol
2013;99:503-509.
9. Lo CW. Role of Gap Junctions in Cardiac Conduction and Development: Insights From the Connexin
EP

Knockout Mice. Circ Res 2000;87:346-348.

10. Von Bibra H, Paulus W, St John Sutton M. Cardiometabolic Syndrome and Increased Risk of Heart
C

Failure. Curr Heart Fail Rep 2016;13:219-229.

AC

11. Menezes AR, Lavie CJ, Dinicolantonio JJ, et al. Cardiometabolic risk factors and atrial fibrillation.
Rev Cardiovasc Med 2013;14:73-81.
12. Carlhäll S, Källén K, Blomberg M. Maternal body mass index and duration of labor. Eur J Obstet
Gynecol Reprod Biol 2013;171:49-53.
13. Sheiner E, Levy A, Feinstein U, et al. Risk factors and outcome of failure to progress during the first
stage of labor: A population-based study. Acta Obstet Gynecol Scand 2002;81:222-26.
14. JanCleary-Goldman J, Malone FD, Vidaver J, et al. Impact of Maternal Age on Obstetric Outcome:
 ACCEPTED MANUSCRIPT

the FASTER trial. Obstet Gynecol 2005;105:983-90.

15. Macones GA. Management of labor and delivery 2015. Chichester, West Sussex: Wiley Blackwell.
16. Arulkumaran S, Regan L. Obstetrics and gynecology 2011. Oxford: Oxford University Press.
17. Nystedt A, Hildingsson I. Diverse definitions of prolonged labour and its consequences with
sometimes subsequent inappropriate treatment. BMC Pregnancy Childbirth 2014;14.

PT
18. Hanley GE, Munro S, Greyson D, et al. Diagnosing onset of labor: a systematic review of definitions
in the research literature. BMC Pregnancy Childbirth 2016;16.

RI
19. Li W, Li H, Long Y. Clinical characteristics and long-term predictors of persistent left ventricular
systolic dysfunction in peripartum cardiomyopathy. Can J Cardiol 2016;32:362-8.

SC
20. Fett JD, McNamara DM. Peripartum cardiomyopathy in 2015. Can J Cardiol 2016;32:286-8.
21. Brown HK, Wilk P. Changes in smoking during pregnancy in Ontario, 1995 to 2010: results from the

U
Canadian community health survey. Journal of Obstetrics and Gynaecology Canada 2014; 36:878-84.
AN
M
D
TE
C EP
AC

14
 ACCEPTED MANUSCRIPT

Table 1. Characteristics of 1,608,720 women at their first delivery during the study, stratified by exposure to prolonged
labour or uterine inertia. All data presented as number (%) unless otherwise indicated.

Exposed to prolonged
Exposed to second or unspecified Unexposed to any
prolonged first stage of labour, or to prolonged labour or
stage of labour uterine inertia uterine inertia
Characteristica (n = 23,318) (n = 240,348) (n = 1,345,054)

PT
At the time of delivery
Mean maternal (SD) age at delivery, years 28.6 (5.5) 29.0 (5.3) 28.6 (5.6)
16 to 19 1,309 (5.6) 11,058 (4.6) 84,621 (6.3)

RI
20 to 24 4,038 (17.3) 36,646 (15.2) 229,454 (17.1)
25 to 29 7,601 (32.6) 80,061 (33.3) 426,805 (31.7)
30 to 34 7,098 (30.4) 76,620 (31.9) 403,937 (30.0)

SC
35 to 39 2,771 (11.9) 30,070 (12.5) 166,810 (12.4)
40 to 44 481 (2.1) 5,648 (2.3) 31,848 (2.4)
45 to 50 20 (0.1) 245 (0.1) 1,579 (0.1)
Income quintile (Q)

U
Q1 (lowest) 6,125 (26.3) 53,091 (22.1) 321,019 (23.9)
Q2 4,885 (20.9) 49,972 (20.8) 281,733 (20.9)
AN
Q3 4,560 (19.6) 48,436 (20.2) 267,008 (19.9)
Q4 4,285 (18.4) 48,979 (20.4) 260,606 (19.4)
Q5 (highest) 3,365 (14.4) 38,825 (16.2) 207,010 (15.4)
M

Unknown 98 (0.4) 1,045 (0.4) 7,678 (0.6)

Rural residence 2,037 (8.7) 22,048 (9.2) 135,545 (10.1)
Previous number of deliveries
0 21,152 (90.7) 222,962 (92.8) 1,074,107 (79.9)
D

1 1,612 (6.9) 13,562 (5.6) 208,303 (15.5)

2 366 (1.6) 2,160 (0.9) 45,177 (3.4)
TE

≥3 108 (0.5) 618 (0.3) 11,662 (0.9)

Unknown 80 (0.3) 1,046 (0.4) 5,805 (0.4)
Mean (SD) duration of hospital stay for delivery,
3.0 (2.1) 3.2 (7.7) 2.8 (2.4)
EP

days
Mean (SD) gestational age at delivery, weeks 38.6 (1.7) 38.3 (1.6) 37.9 (1.8)
Gestational age at delivery, weeks
24-28 25 (0.1) 83 (0.0) 6,610 (0.5)
C

29-32 69 (0.3) 272 (0.1) 10,629 (0.8)

33-36 579 (2.5) 6,104 (2.5) 74,143 (5.5)
AC

≥ 37 22,645 (97.1) 233,889 (97.3) 1,253,672 (93.2)

Mean (SD) infant birthweight, grams 3,481.8 (491.2) 3,527.1 (495.3 3,331.2 (569.9)
Infant birthweight category, grams
< 1000 15 (0.1) 68 (0.0) 5,522 (0.4)
1000-1499 45 (0.2) 181 (0.1) 7,687 (0.6)
1500-2499 486 (2.1) 4,383 (1.8) 70,168 (5.2)
2500-3999 19,666 (84.3) 196,412 (81.7) 1,127,095 (83.8)
≥ 4000 3,106 (13.3) 39,304 (16.4) 134,582 (10.0)
 ACCEPTED MANUSCRIPT

Exposed to prolonged
Exposed to second or unspecified Unexposed to any
prolonged first stage of labour, or to prolonged labour or
stage of labour uterine inertia uterine inertia
Characteristica (n = 23,318) (n = 240,348) (n = 1,345,054)
Conditions ≤ 24 months before, or at, the
delivery date
Drug dependence or tobacco use 490 (2.1) 4,842 (2.0) 27,463 (2.0)

PT
Obesity 800 (3.4) 8,984 (3.7) 40,662 (3.0)
Diabetes mellitus 1,366 (5.9) 16,399 (6.8) 87,003 (6.5)
Chronic hypertension 880 (3.8) 11,147 (4.6) 60,810 (4.5)

RI
Kidney disease 84 (0.4) 708 (0.3) 4,457 (0.3)
Dyslipidemia 424 (1.8) 4,502 (1.9) 23,876 (1.8)

SC
Preeclampsia/eclampsia at the delivery
350 (1.5) 6,364 (2.6) 36,147 (2.7)
hospitalization or ≤ 42 days postpartum

Characteristics of the labour and delivery

U
Type of labour
Prolonged first stage of labour 23,318 (100.0) 0 (0.0) 0 (0.0)
AN
Prolonged second stage of labour 3,235 (13.9) 76,891 (32.0) 0 (0.0)
Any uterine inertia 4,290 (18.4) 166,179 (69.1) 0 (0.0)
Previous Caesarean section (parous only) 327 (15.7) 3,252 (19.9) 54,617 (20.6)
M

Caesarean section at the index delivery 5,943 (25.5) 97,168 (40.4) 292,451 (21.7)
Disproportion or obstruction due to bony pelvis 1,293 (5.5) 11,985 (5.0) 30,341 (2.3)
Shoulder dystocia 583 (2.5) 5,042 (2.1) 22,949 (1.7)
D

Other obstructed labour 1,094 (4.7) 10,414 (4.3) 27,558 (2.0)

Fetal malposition or presentation 2,504 (10.7) 28,148 (11.7) 135,811 (10.1)
Forceps or vacuum (vaginal deliveries) 5,505 (23.6) 76,317 (31.8) 201,149 (15.0)
TE

Induced labour 4,991 (21.4) 73,097 (30.4) 297,384 (22.1)

Study duration of follow-up

EP

Median (IQR) duration of follow-up, years 10.5 (5.7 to 17.0) 12.8 (6.9 to 18.5) 14.0 (7.6 to 20.4)

a
Includes each woman's first delivery during the study period, regardless of whether she was exposed to prolonged
labour at a subsequent delivery.
C
AC
 ACCEPTED MANUSCRIPT

Table 2 (main model). Future risk of hospitalization for newly diagnosed heart failure, cardiomyopathy or cardiac dysrhythmia starting 1 day
after the index delivery discharge date, in association with exposure to prolonged first stage of labour vs. no prolonged labour or uterine
inertia at any delivery during the study.

Exposed to prolonged Unexposed to any prolonged labour or

PT
first stage of labour uterine inertia Hazard ratio
(n = 31,150 women) (n = 1,337,222 women) (95% confidence interval)

RI
No. events Incidence rate per No. events Incidence rate per
Outcome (%) 10,000 person-years (%) 10,000 person-years Crude Adjusteda

SC
Primary composite 78 (0.15) 2.33 4114 (0.18) 2.30 1.07 (0.86 to 1.34) 1.09 (0.87 to 1.36)
of hospitalization for
newly diagnosed

U
heart failure,

AN
cardiomyopathy or
cardiac dysrhythmia

M
Individual outcomes

D
Hospitalization for 46 (0.09) 1.37 2343 (0.10) 1.31 1.11 (0.80 to 1.49) 1.13 (0.84 to 1.51)

TE
newly diagnosed
heart failure or
cardiomyopathy
EP
Hospitalization for 33 (0.06) 0.99 2177 (0.10) 1.22 0.86 (0.61 to 1.21) 0.87 (0.62 to 1.23)
newly diagnosed
C

cardiac dysrhythmia
AC

a
Adjusted for maternal age, parity, obstructed labour or fetal malposition, preeclampsia, income quintile, rural residence, preterm birth
and infant birthweight -- each at the time of a given delivery; time-varying drug/tobacco use, obesity, diabetes mellitus, chronic hypertension,
kidney disease, dyslipidemia -- each diagnosed before or at the time of a given delivery; as well as newly diagnosed coronary artery disease or
congenital heart disease arising at least 1 day after the index delivery discharge date.
 ACCEPTED MANUSCRIPT

Table 3. Future risk of hospitalization for newly diagnosed heart failure, cardiomyopathy or cardiac dysrhythmia starting 1 day after the index
delivery discharge date, in association with type of prolonged labour or uterine inertia vs. no prolonged labour or uterine inertia of any type
at any delivery during the study.

PT
Incidence rate per Crude hazard ratio Adjusted hazard ratio
Type of prolonged laboura No. events (%) 10,000 person-years (95% confidence interval) (95% confidence interval)b
Prolonged first stage 78 (0.15) 2.33 1.07 (0.86 to 1.34) 1.09 (0.87 to 1.36)

RI
(main exposure)

SC
Prolonged second stage 212 (0.13) 1.86 0.85 (0.74 to 0.97) 0.86 (0.75 to 0.99)
(secondary exposure #2)

U
Any prolonged labour and/or 782 (0.15) 2.05 0.92 (0.86 to 1.00) 0.92 (0.85 to 0.89)

AN
uterine inertia (secondary
exposure #3)

M
Any uterine inertia 528 (0.15) 2.12 0.96 (0.88 to 1.05) 0.94 (0.86 to 1.03)
(secondary exposure #4)

D
TE
a
Each model was restricted to women with any exposure to a given type prolonged labour and women with no exposure to prolonged labour or uterine inertia
of any type (the referent), at the time of a given delivery.
EP
b
Adjusted for maternal age, parity, obstructed labour or fetal malposition, preeclampsia, income quintile, rural residence, preterm birth and infant
birthweight -- each at the time of a given delivery; time-varying drug/tobacco use, obesity, diabetes mellitus, chronic hypertension, kidney disease,
dyslipidemia -- each diagnosed before or at the time of a given delivery; as well as newly diagnosed coronary artery disease or congenital heart disease arising
C

at least 1 day after the index delivery discharge date.

AC
 ACCEPTED
Figure 1 (additional analysis 2). Incidence rate and adjusted MANUSCRIPT
hazard ratios for the composite outcome of heart failure,
cardiomyopathy or cardiac arrhythmia starting 1 day after the index delivery discharge date, in association with exposure
to prolonged first stage of labour. Data are stratified by parity, prior caesarean delivery, as well as current spontaneous
vaginal delivery, induction of labour and forceps or vacuum assisted vaginal delivery.

No. outcome events

(rate per 10,000 person-years)

Nulliparous: 29 (2.34)
Parity

PT
Parous: 49 (2.33)

RI
SC
Prior Caesarean delivery
among parous women

No: 28 (2.28)

Yes: < 6 (6.11)

U
AN
M
Current spontaneous
vaginal delivery

32 (2.02)
D
TE
Current induc on
of labour

EP

20 (3.05)
C
Current forceps or vacuum
assisted vaginal delivery

AC

12 (2.41)

0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 8.0 9.0 10.0 11.0

Adjusted hazard ratio (95% CI)*

*Adjusted for maternal age, parity, obstructed labour or fetal malposition, preeclampsia, income quintile, rural residence, preterm
birth and infant birthweight -- each at the time of a given delivery; time-varying drug/tobacco use, obesity, diabetes mellitus,
chronic hypertension, kidney disease, dyslipidemia -- each diagnosed before or at the time of a given delivery; as well as newly
diagnosed coronary artery disease or congenital heart disease arising at least 1 day after the index delivery discharge date.