- Email: [email protected]
Influence of paternal age on perinatal outcomes
Accepted Manuscript Influence of Paternal Age on Perinatal Outcomes Emily G. Hurley, MD, Emily A. DeFranco, DO, MS PII:
S0002-9378(17)30925-0
DOI:
10.1016/j.ajog.2017.07.034
Reference:
YMOB 11789
To appear in:
American Journal of Obstetrics and Gynecology
Received Date: 23 March 2017 Revised Date:
19 July 2017
Accepted Date: 31 July 2017
Please cite this article as: Hurley EG, DeFranco EA, Influence of Paternal Age on Perinatal Outcomes, American Journal of Obstetrics and Gynecology (2017), doi: 10.1016/j.ajog.2017.07.034. 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 1 1
INFLUENCE OF PATERNAL AGE ON PERINATAL OUTCOMES.
2
Emily G. HURLEY, MDa, Emily A. DeFRANCO, DO, MSa,b
3
a
4 5
b
Center for Prevention of Preterm Birth, Perinatal Institute, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio
6 7
RI PT
University of Cincinnati, Department of Obstetrics and Gynecology, Cincinnati, Ohio
Disclosure Statement: The authors report no conflict of interest.
Presentation information: The abstract for this study was presented as a poster presentation at the American Society for Reproductive Medicine Scientific Meeting October 15-19, 2016 in Salt Lake City, Utah, USA.
11 12 13 14
Funding: This work was supported by the Perinatal Institute, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA, and March of Dimes Grant 22-FY14-470 for the March of Dimes Prematurity Research Center Ohio, Collaborative, USA.
15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
Corresponding author: Emily G. Hurley, MD University of Cincinnati College of Medicine Department of Obstetrics and Gynecology 231 Albert Sabin Way, Mail Location 0526 Cincinnati, Ohio 45267-0526 United States of America Cell phone: 1-513-919-7696 Fax: 1-513-558-3558 Email: [email protected]
32
AC C
Word count: Abstract 333 Main text 2,117
EP
TE D
M AN U
SC
8 9 10
Table 3 is to appear in the print issue should the submission be accepted for publication.
ACCEPTED MANUSCRIPT 2 33
Condensation: Older paternal age does not increase risk of adverse perinatal outcomes.
34
Short version of article title: PATERNAL AGE AND PERINATAL OUTCOMES
RI PT
35 36
AC C
EP
TE D
M AN U
SC
37
ACCEPTED MANUSCRIPT 3
Abstract
39
Background: There is an increasing trend to delay childbearing to advanced parental age.
40
Increased risks of advanced maternal age and assisted reproductive technologies are widely
41
accepted. There is limited data regarding advanced paternal age. In order to adequately
42
counsel patients on risk, more research regarding advanced paternal age is necessary.
43
Objective: To determine the influence of paternal age on perinatal outcomes, and to assess
44
whether this influence differs between pregnancies achieved spontaneously and those
45
achieved with assisted reproductive technology (ART).
46
Study Design: A population-based retrospective cohort study of all live births in Ohio from
47
2006-2012 was completed. Data were evaluated to determine if advanced paternal age is
48
associated with an increased risk of adverse outcomes in pregnancies. The analysis was
49
stratified by status of utilization of ART. Generalized linear regression models assessed the
50
association of paternal age on pregnancy complications in ART and spontaneously conceived
51
pregnancies, after adjusting for maternal age, race, multifetal gestation and Medicaid status,
52
using STATA software.
53
Results: Paternal age was documented in 82.2% of 1,034,552 live births in Ohio during the 7-
54
year study period. Paternal age ranged from 12-87 years, median 30 (IQR 26, 35). Maternal
55
age ranged from 11-62 years, median 27 (IQR 22, 31). The use of ART in live births increased as
56
paternal age increased: 0.1% <30 years vs. 2.5% > 60 years, p<0.001. After accounting for
57
maternal age and other confounding risk factors, increased paternal age was not associated
58
with a significant increase in the rate of pre-eclampsia, preterm birth, fetal growth restriction,
59
congenital anomaly, genetic disorder or Neonatal Intensive Care Unit admission. The influence
AC C
EP
TE D
M AN U
SC
RI PT
38
ACCEPTED MANUSCRIPT 4
of paternal age on pregnancy outcomes was similar in pregnancies achieved with ART
61
compared to non-ART pregnancies.
62
Conclusion: Older paternal age does not appear to pose an independent risk of adverse
63
perinatal outcomes, either in pregnancies achieved with or without ART. However, small effect
64
sizes such as very small risk increases or decreases may not be detectable despite the large
65
sample size in this study of over 830,000 births.
SC
66
Keywords: Advanced paternal age; assisted reproductive technology; perinatal outcomes
M AN U
67
RI PT
60
68
AC C
EP
TE D
69
ACCEPTED MANUSCRIPT 5 70 71
Introduction There is an increasing trend to delay childbearing to advanced parental age, mother and father alike. The associated risks of advanced maternal age, generally defined ≥35 years of
73
age, are well studied. Increased maternal age is associated with increased miscarriage rate,
74
obstetric complications and perinatal mortality and is also associated with decreased success
75
when using assisted reproductive technologies (ART).1-2 Less is known regarding the risks
76
associated with advanced paternal age and pregnancy outcomes. Some studies have shown
77
that advanced paternal age has effects on sperm quality, miscarriage rates, infant birth weights,
78
and childhood/adult illnesses.1,3-8 Nonetheless, one study published in 2008 by Chen et al.
79
found that advanced paternal age is not associated with adverse birth outcomes.9
SC
M AN U
80
RI PT
72
As parental ages increase, more couples may turn to ART for aid in achieving pregnancy. ART itself is associated with increased perinatal risks including prematurity, low birth weight
82
and multiple gestations.10 However, there is a paucity of data evaluating the impact of
83
advanced paternal age on the risks associated with ART pregnancies. A review of 10 articles
84
published in 2011 found that there is insufficient evidence available to determine if paternal
85
age has a negative impact on ART outcomes. The outcomes evaluated included fertilization
86
rate, pregnancy rate, miscarriage rate and live birth rate. Only four of the studies reviewed
87
evaluated live birth rate with two of the studies finding no significant difference and two
88
studies with a decreased chance of live birth with advanced paternal age.1 A study published in
89
2014 found no increased risk for prematurity, low birthweight or small for gestational age in
90
pregnancies achieved with ART at older paternal age.11 After a thorough review of the
AC C
EP
TE D
81
ACCEPTED MANUSCRIPT 6 91
literature, overall there is a limited amount of data regarding advanced paternal age and
92
perinatal outcomes in pregnancies achieved with ART. The aim of this study was to utilize a large population-based data source to assess
RI PT
93
whether advanced paternal age poses independent increased perinatal risks, and assess
95
whether those risks differ in pregnancies achieved by ART or those achieved spontaneously
96
without ART.
97
Material and Methods
99
A population-based retrospective cohort study of all live births in Ohio from 2006-2012
M AN U
98
SC
94
was performed to quantify the influence of advanced paternal age on pregnancy complications. Study approval was obtained from the Ohio Department of Health Institutional Review Board
101
and a de-identified data set extracted from live-birth certificates was provided for this analysis.
102
This study was exempt from review by the Institutional Review Board at the University of
103
Cincinnati, Cincinnati, Ohio.
104
TE D
100
The exposure of interest was paternal age. Paternal age was evaluated both continuously and categorically (<30, 30-39, 40-49, 50-59 and >60 years of age). The 2003
106
version of the birth certificate includes three areas to indicate use of fertility treatment under
107
the header “RISK FACTORS IN THIS PREGNANCY.” The initial option included is pregnancy
108
resulting from infertility treatment, and if chosen, the two further options include 1.) use of
109
fertility-enhancing drugs, artificial insemination or intrauterine insemination and 2.) assisted
110
reproductive technology (in vitro fertilization, gamete intrafallopian transfer).12 Guidelines and
111
instructions for identification of use of infertility treatments resulting in live births in the US is
112
outlined in the National Center for Health Statistics Guide to Completing the Facility
AC C
EP
105
ACCEPTED MANUSCRIPT 7
Worksheets for the Certificate of Live Birth and Report of Fetal Death (2003 revision).13 For this
114
study, ART status was assigned as “assisted reproductive technology” recorded on page 2, in
115
field number 41 on the live birth certificate.12
116
RI PT
113
Outcomes of interest for this study included: pre-eclampsia, preterm birth, fetal growth restriction, congenital anomaly, genetic disorder, and Neonatal Intensive Care Unit (NICU)
118
admission. Pre-eclampsia included cases recorded on the birth certificate as gestational
119
hypertension, pregnancy induced hypertension (PIH), or pre-eclampsia. Preterm birth was
120
defined as gestational age at time of birth recorded as <37 weeks by obstetric estimate, which
121
includes data from both last menstrual period and ultrasound estimates of gestational age.
122
Fetal growth restriction was defined as birthweight <10th percentile using a large US national
123
growth reference.14 Any major congenital anomaly included newborns with anencephaly,
124
meningomyelocele/spina bifida, congenital heart disease, congenital diaphragmatic hernia,
125
omphalocele, gastroschisis, limb reduction defect, cleft lip and/or cleft palate, or hypospadias
126
indicated in field number 55 of the birth certificate. Genetic disorder included cases of Down
127
syndrome or other suspected chromosomal disorder at the time of birth certificate recording.
128
NICU admission included newborns admitted to the Neonatal Intensive Care Unit during the
129
first few days of life, again as recorded at the time of birth certificate generation.12
M AN U
TE D
EP
AC C
130
SC
117
To quantify the effect of paternal age on the dichotomous outcomes of adverse
131
perinatal outcomes including pre-eclampsia, preterm birth, fetal growth restriction, congenital
132
anomaly, genetic disorder, and NICU admission, we used a log-binomial model, Generalized
133
Linear Model (GLM), with log link. This linear model approach for estimating relative risk, is
134
well-suited for dichotomous outcomes. Risk estimates for pregnancy complications were
ACCEPTED MANUSCRIPT 8
stratified by ART and non-ART pregnancies. Adjusted relative risk (RR) values and 95%
136
confidence intervals were calculated to estimate the risk of advancing paternal age on each
137
adverse pregnancy outcome after adjustment for the confounding influences of advancing
138
maternal age, race, Medicaid status, and multifetal gestation. Sensitivity analyses were
139
performed estimating relative risk with maternal age held constant, and also imputing missing
140
data on paternal age as a spectrum of ages. Statistical analysis was performed using STATA
141
Release 12 Software (StataCorp, College Station, TX, USA).
142
Results
SC
M AN U
143
RI PT
135
Paternal age was documented in approximately 82.2% of 1,034,552 live births in Ohio between the years of 2006-2012. A total of 833,727 births were included in the study after
145
excluding those without a paternal age recorded and limiting to the first birth of multifetal
146
gestations. Approximately 3.7% of the births (3,118) were achieved with ART compared to
147
830,609 non-ART births. Paternal age ranged from 12-87 years, median 30 (IQR 26, 35).
148
Maternal age ranged from 11-62 years, median 27 (IQR 22, 31). As paternal age increased, the
149
corresponding maternal age also increased, however not linearly. When mothers were <20
150
years of age, the majority (96.2%) of fathers were <30 years of age. When the paternal age was
151
>40, the majority of mothers were >35 years of age (Table 1).
EP
AC C
152
TE D
144
Parental demographic characteristics of the live birth population are displayed in Table
153
1. There were higher rates of both mothers and fathers with less than a high school education
154
when the paternal age was in the youngest (<30 years) and oldest (>60) categories. Older
155
paternal age was associated with a higher proportional frequency of non-Hispanic black
ACCEPTED MANUSCRIPT 9 156
maternal race. Utilization of state-funded Medicaid insurance and maternal tobacco use were
157
higher in pregnancies with young and old paternal age. The frequency of perinatal complications was higher in the youngest and oldest paternal
RI PT
158
age groups, as displayed in Table 2. In unadjusted analysis, prior to accounting for the
160
confounding risk of maternal age, the crude risk of pregnancy complications was highest in the
161
extreme categories of paternal age. The paternal age category with the lowest frequency and
162
unadjusted risk of pregnancy complications was 30-39 years of age (Table 2). However,
163
adjustment for maternal age resulted in null independent influence of paternal age on adverse
164
pregnancy outcomes, indicating that the observed influence of paternal age was attributable
165
primarily to concordant similar extremes of maternal age (Table 3). These findings were similar
166
in pregnancies utilizing ART and those conceived without the assistance of ART. Sensitivity
167
analysis performed removing maternal age from the model and holding maternal age constant,
168
and also imputing missing data on paternal age as a variety of ages did not significantly alter the
169
relative risk effect estimates displayed in Table 3.
M AN U
TE D
The use of ART in live births significantly increased as paternal age increased with 0.1%
EP
170
SC
159
of pregnancies achieved with ART when paternal age was <30 compared to 2.5% when >60
172
years of age (p<0.001) (Table 4). There was a significant interaction between the oldest
173
paternal age category, ≥60 years and ART (p=0.02), and stratified analyses by ART status were
174
performed. After accounting for the confounding influences of maternal age, race, Medicaid
175
status and multifetal gestation, increased paternal age was not associated with a significant
176
increase in the rate of pre-eclampsia, preterm birth, fetal growth restriction, congenital
AC C
171
ACCEPTED MANUSCRIPT 10
anomaly, genetic disorder or NICU admission with analyses stratified by use of ART, with all
178
adjusted relative risk estimate 95% confidence intervals crossing the null value of 1 (Table 3).
179
Comment
RI PT
177
As couples delay childbearing, the question of the effects of older age gain importance.
181
Advanced maternal age is well defined with understood risks and implications, however, this is
182
not true for advanced paternal age. There is no widely accepted definition of advanced
183
paternal age and the consequences of pregnancies achieved at advanced paternal age are not
184
well understood. Also, with advanced parental age, couples are more likely to utilize ART which
185
has risks of its own. Our study demonstrates that advanced paternal age does not increase the
186
risk of poor perinatal outcomes, even when stratified by pregnancies achieved with ART
187
compared to spontaneous pregnancies. After adjusting for maternal age and other
188
confounding risk factors, the risk of pre-eclampsia, preterm birth, fetal growth restriction,
189
congenital anomaly, genetic disorder or NICU admission was not increased with advancing
190
paternal age.
EP
TE D
M AN U
SC
180
Our study is consistent with a previous study completed by Stern et al., which was
192
published in 2014. The study was smaller (n=280,804) and only evaluated the effect of paternal
193
age on three outcomes including prematurity, low birthweight and small for gestational age.
194
The outcomes were limited to mothers aged 40 and younger. Nonetheless, the study found
195
that older paternal age was not associated with increased risk for prematurity, low birthweight
196
or small for gestational age in pregnancies in fertile, subfertile and infertile populations using
197
ART.11
AC C
191
ACCEPTED MANUSCRIPT 11
A previous population-based retrospective cohort study published by Yang et al. found
199
that older paternal age is associated with an increased risk of birth defects. The study by Yang
200
and colleagues had a much larger sample size with over 5 million subjects, a sample size larger
201
than six times the 833,727 births included in our study. 15 Our study also identified an increased
202
relative risk of congenital anomalies with increasing paternal age similar to the effect size
203
identified in the study by Yang (Table 3), however it was not statistically significant. The
204
absolute risk increase observed by Yang was only 0.4% increased prevalence, therefore the
205
significance of the result may not be clinically meaningful and may represent alpha error. The
206
difference between the two studies may be a reflection of the difference in the sample sizes
207
however is not a difference in effect size.
M AN U
SC
RI PT
198
Limitations to this study include the retrospective use of birth certificate data, both in
209
regards to paternal age and use of ART. Although there are published guidelines and training
210
on how to fill out the US birth certificate, the accuracy of recorded variables may vary.
211
Approximately 17.8% of births recorded in Ohio from 2006-2012 did not have paternal age
212
recorded, which may reflect unknown paternity. Also we were unable to assess whether the
213
recorded paternal age took into consideration whether a sperm donor was used. A study
214
published in July 2016 compared the Society for Assisted Reproductive Technology Clinic Online
215
Reporting System to birth certificate data of 8 states and found that only 36.5% of children
216
conceived utilizing in vitro fertilization was accurately reported on the birth certificate.16 We
217
would presume that those recorded on the birth certificate as ART were accurate, however
218
there may be a significant number of ART cases listed as no ART, or missing data for that field.
219
Thus, our results may underestimate the risks in ART pregnancies as some ART pregnancies may
AC C
EP
TE D
208
ACCEPTED MANUSCRIPT 12
be included in the referent group. However, we presume this influence would be minimal as no
221
significant risk increase for outcomes was observed overall, or in the stratified analysis among
222
ART and non-ART pregnancies. Furthermore, sensitivity analyses modeling missing data on
223
paternal age as a variety of young, mid-range, and older paternal ages did not significantly alter
224
our observed relative risk estimates.
Another limitation of this study is that long-term outcomes cannot be evaluated. Given
SC
225
RI PT
220
that our findings are based on birth certificate data, we were unable to assess later diagnosed
227
childhood or adult illness of infants born to fathers of older age. Also, this study is limited to
228
the state of Ohio, which has a population that may not be generalizable to all pregnancies
229
across the United States and in other countries. In addition, as in any observational study,
230
unidentified and unmeasured confounds exist which are unable to be accounted for. A major
231
strength of this study was the large sample size.
TE D
M AN U
226
In summary, we found that advanced paternal age is not significantly associated with an
233
independent increase in risk of adverse perinatal outcomes. The observed increased frequency
234
of adverse outcomes at the extremes of paternal age became non-significant after adjusting for
235
the confounding influence of maternal age and other risk factors. These findings suggest that
236
maternal age is the primary contributor to adverse outcomes in couples who delay childbearing
237
to older parental ages. This information may be clinically useful when counseling couples
238
planning on parental age-related pregnancy risks.
239 240
AC C
EP
232
ACCEPTED MANUSCRIPT 13
Acknowledgements
242
This study includes data provided by the Ohio Department of Health, which should not be
243
considered an endorsement of this study or its conclusions.
RI PT
241
244 245
SC
246
M AN U
247 248
AC C
EP
TE D
249
ACCEPTED MANUSCRIPT 14
251
References 1.
outcome. Fertil Steril. 2011;95(1):1–8.
252 253
Dain L, Auslander R, Dirnfeld M. The effect of paternal age on assisted reproduction
2.
RI PT
250
Valent AM, Newman T, Chen A, Thompson A, Defranco E. Gestational age-specific neonatal morbidity among pregnancies complicated by advanced maternal age: A
255
population-based retrospective cohort study. J Matern Neonatal Med.
256
2016;29(9):1485–90. 3.
Sharma R, Agarwal A, Rohra V, Assidi M, Abu-Elmagd M, Turki RF. Effects of
M AN U
257
SC
254
258
increased paternal age on sperm quality, reproductive outcome and associated
259
epigenetic risks to offspring. Reprod Biol Endocrinol. 2015;13(1):35.
260
4.
de la Rochebrochard E, Thonneau P. Paternal age and maternal age are risk factors for miscarriage; results of a multicentre European study. Hum Reprod.
262
2002;17(6):1649–56. 5.
6.
7.
270
Reichman NE, Teitler JO. Paternal age as a risk factor for low birthweight. Am J Public
Health. 2006;96(5):862–6.
268 269
Slama R, Bouyer J, Windham G, Fenster L, Werwatz A, Swan SH. Influence of paternal age on the risk of spontaneous abortion. Am J Epidemiol. 2005;161(9):816–23.
266 267
EP
and spontaneous abortion. Obstet Gynecol. 2006;108(2):369–77.
264 265
Kleinhaus K, Perrin M, Friedlander Y, Paltiel O, Malaspina D, Harlap S. Paternal age
AC C
263
TE D
261
8.
Ramasamy R, Chiba K, Butler P, Lamb DJ. Male biological clock: a critical analysis of advanced paternal age. Fertil Steril. 2015;103(6):1402–6.
ACCEPTED MANUSCRIPT 15
9.
Chen X, Wen S, Krewski D, Fleming N, Yang Q, Walker M. Paternal age and adverse birth outcomes: Teenager or 40+, who is at risk? Hum Reprod. 2008;23(6):1290–6.
272 273
10.
reproductive technology. Obstet Gynecol. 2005;106(5 Pt 1):1143–6.
274 275
ACOG Committee Opinion #324: Perinatal risks associated with assisted
RI PT
271
11.
Stern JE, Luke B, Hornstein MD, et al. The effect of father’s age in fertile, subfertile, and assisted reproductive technology pregnancies: A population based cohort study.
277
J Assist Reprod Genet. 2014;31(11):1437–44. 12.
http://www.cdc.gov/nchs/data/dvs/birth11-03final-ACC.pdf.
279 280
U.S. Standard Certificate of Live Birth. Revised 11/2003. Available from:
M AN U
278
SC
276
13.
National Center for Health Statistics. Guide to Completing the Facility Worksheets for the Certificate of Live Birth and Report of Fetal Death (2003 revision). Hyattsville,
282
MD. US Department of Health and Human Services, Centers for Disease Control and
283
Prevention, 2016. Available from:
284
http://www.cdc.gov/nchs/data/dvs/GuidetoCompleteFacilityWks.pdf. 14.
reference for fetal growth. Vol. 87, Obstetrics and gynecology. 1996. p. 163–8.
286
288 289
15.
Yang Q, Wen SW, Leader A, Chen XK, Lipson J, Walker M. Paternal age and birth
AC C
287
Alexander GR, Himes JH, Kaufman RB, Mor J, Kogan M. A United States national
EP
285
TE D
281
defects: how strong is the association? Human Reproduction. 2007; 22: 696-701.
16.
Luke BB, Brown MB, Spector LG. Validation of infertility treatment and assisted
290
reproductive technology use on the birth certificate in eight states. Vol. 215. Am J
291
Obstet Gynecol. 2016. p. 126-127.
292
ACCEPTED MANUSCRIPT 16 293
Table 1. Baseline characteristics stratified by paternal age categories.
Characteristics
Paternal Age
p-value
30-39
40-49
50-59
≥ 60
N=373,649
N=379,942
N=73,075
N=6,459
N=602
(36.8%)
(37.4%)
(7.2%)
(0.6%)
(0.06%)
<20
56,844 (15.2%)
1,954 (0.5%)
223 (0.3%)
35 (0.5%)
3 (0.5%)
20-34
311,982(83.5%) 314,813 (82.9%)
31,188 (42.7%)
2,737 (42.1%)
255 (42.4%)
>35
3,408 (0.9%)
41,420 (56.7%)
3,655 (56.6%)
342 (56.8%)
M AN U
62,010 (16.3%)
SC
Maternal age
RI PT
<30
White
301,707(80.7%) 320,893 (84.5%)
58,093 (79.5%)
4,465 (69.1%)
353 (58.6%)
Black
46,055 (12.3%)
29,307 (7.7%)
8,766 (12%)
1,273 (19.7%)
158 (26.2%)
Hispanic
20,554 (5.5%)
13,861 (3.6%)
2,609 (3.6%)
281 (4.3%)
20 (3.3%)
Other
5,333 (1.4%)
TE D
Maternal Race/Ethnicity
15,881 (4.2%)
3,607 (4.9%)
440 (6.8%)
71 (11.8%)
<0.001
<0.001
58,791 (15.7%)
21,659 (5.7%)
4,333 (5.9%)
507 (7.8%)
66 (11.0%)
<0.001
Maternal
71,650 (19.2%)
28,332 (7.5%)
6,554 (9.0%)
860 (13.3%)
114 (18.9%)
<0.001
Maternal Tobacco Use
103,541(27.7%) 60,822 (16%)
11,699 (16%)
1,408 (21.8%)
122 (20.3%)
<0.001
Maternal Medicaid use
164,045(42.9%) 69,287 (18.2%)
14,556 (19.9%)
1,945 (30.1%)
202 (33.5%)
<0.001
Mother Unmarried
178,702(47.8%) 62,979 (16.6%)
12,374 (16.9%)
1,554 (24.1%)
139 (23.1%)
<0.001
AC C
294 295 296 297
EP
Paternal
ACCEPTED MANUSCRIPT 17 Table 2. Pregnancy complications and delivery characteristics stratified by paternal age.
Maternal-fetal complication
Fetal growth restriction RR (95% CI) Major fetal anomaly RR (95% CI) Genetic disorder RR (95% CI)
299 300
≥ 60
Missing age
N=373,649
N=379,942
N=73,075
N=6,459
N=602
N=180,841
(36.8%)
(37.4%)
(7.2%)
(0.6%)
(0.06%)
19,337 (5.2%)
17,970 (4.8%)
3,814 (5.3%)
362 (5.7%)
32 (5.4%)
9701 (5.5%)
1.06 (1.04,1.08)
0.92 (0.90,0.94)
1.07 (1.04,1.10)
1.15 (1.04,1.27)
1.12 (0.82,1.54)
1.09 (1.07,1.12)
40,495 (10.8%)
38,284 (10.1%)
8,999 (12.3%)
959 (14.8%)
86 (14.3%)
28,838 (16.6%)
1.00 (0.99,1.01)
0.92 (0.92,0.94)
1.19 (1.17,1.21)
1.40 (1.33,1.48)
1.38 (1.16,1.65)
1.44 (1.43,1.46)
37,370 (10%)
29,750 (7.8%)
6,390 (8.7%)
704 (10.9%)
72 (12.0%)
26,913 (15.5%)
1.22 (1.20,1.24)
0.81 (0.80,0.82)
1.00 (0.98,1.02)
1.23 (1.15,1.31)
1.35 (1.09,1.65)
1.65 (1.63,1.67)
1,316 (0.35%)
1,085 (0.29%)
238 (0.33%)
17 (0.26%)
2 (0.33%)
691 (0.40%)
1.20 (1.11,1.29)
0.82 (0.76,0.89)
1.05 (0.92,1.20)
0.85 (0.53,1.34)
0.99 (0.25,3.93)
1.22 (1.12,1.32)
242 (0.06%)
373 (0.10%)
178 (0.24%)
24 (0.37%)
3 (0.50%)
184 (0.10%)
0.52 (0.44,0.60)
1.00 (0.87,1.15)
2.84 (2.41,3.35)
3.76 (2.51,5.65)
4.84 (1.56,15.0)
1.04 (0.89,1.23)
RI PT
50-59
M AN U
RR (95% CI)
40-49
TE D
Preterm birth <37 weeks
30-39
EP
RR (95% CI)
p-value
<30
AC C
Pre-eclampsia
Paternal Age
SC
298
<0.001
<0.001
<0.001
<0.001
<0.001
ACCEPTED MANUSCRIPT 18 Table 3. Influence of increasing paternal age on the risk of adverse pregnancy outcomes in ART and non-ART pregnancies. Non-ART Pregnancies
N=3,118
N=830,609
adjRR (95% CI)*
Crude RR
adjRR (95% CI)*
Pre-eclampsia
1.01 (0.99, 1.03)
1.01 (0.98, 1.03)
1.01 (1.00, 1.01)
1.00 (0.99 ,1.00)
Preterm birth <37 weeks
1.00 (1.02, 1.02)
1.00 (0.99, 1.01)
1.02 (1.02, 1.02)
1.01 (1.01, 1.01)
Fetal growth restriction 0.99 (0.97, 1.01) (birthweight <10th percentile)
1.00 (0.98, 1.03)
Congenital anomaly (any)
1.00 (0.94, 1.07)
0.99 (0.88, 1.11)
Genetic disorder
1.08 (0.94, 1.24)
NICU admission
1.00 (0.98, 1.01)
1.01 (1.01, 1.01)
1.01 (1.01, 1.02)
1.01 (1.00, 1.02)
1.01 (0.99, 1.03)
1.06 (0.88, 1.27)
1.09 (1.07, 1.10)
1.01 (0.99, 1.04)
1.00 (0.98, 1.02)
1.02 (1.02, 1.02)
1.01 (1.01, 1.01)
EP
TE D
*Relative risk limited to paternal age ≥ 30 years, adjusted for maternal age, multifetal birth, maternal race, and Medicaid status, (95% confidence interval).
AC C
303 304 305 306 307 308
SC
Crude RR
M AN U
Adverse Pregnancy Outcome
ART Pregnancies
RI PT
301 302
ACCEPTED MANUSCRIPT 19 Table 4. Use of infertility treatment stratified by paternal age categories. Infertility Treatment Paternal Age 40-49
50-59
≥ 60
N=373,649
N=379,942
N=73,075
N=6,459
N=602
(36.8%)
(37.4%)
(7.2%)
(0.6%)
(0.06%)
1,271
5,210
1,533
151
22
(0.34%)
(1.38%)
(2.11%)
(2.36%)
(3.67%)
428
1,912
667
(0.12%)
(0.51%)
(0.92%)
96
15
(1.5%)
(2.5%)
M AN U
Assisted reproductive technologies
30-39
RI PT
Infertility Drugs and/or insemination
p-value
<30
SC
309
AC C
EP
TE D
310
<0.001
<0.001
S0002-9378(17)30925-0
DOI:
10.1016/j.ajog.2017.07.034
Reference:
YMOB 11789
To appear in:
American Journal of Obstetrics and Gynecology
Received Date: 23 March 2017 Revised Date:
19 July 2017
Accepted Date: 31 July 2017
Please cite this article as: Hurley EG, DeFranco EA, Influence of Paternal Age on Perinatal Outcomes, American Journal of Obstetrics and Gynecology (2017), doi: 10.1016/j.ajog.2017.07.034. 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 1 1
INFLUENCE OF PATERNAL AGE ON PERINATAL OUTCOMES.
2
Emily G. HURLEY, MDa, Emily A. DeFRANCO, DO, MSa,b
3
a
4 5
b
Center for Prevention of Preterm Birth, Perinatal Institute, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio
6 7
RI PT
University of Cincinnati, Department of Obstetrics and Gynecology, Cincinnati, Ohio
Disclosure Statement: The authors report no conflict of interest.
Presentation information: The abstract for this study was presented as a poster presentation at the American Society for Reproductive Medicine Scientific Meeting October 15-19, 2016 in Salt Lake City, Utah, USA.
11 12 13 14
Funding: This work was supported by the Perinatal Institute, Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio, USA, and March of Dimes Grant 22-FY14-470 for the March of Dimes Prematurity Research Center Ohio, Collaborative, USA.
15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
Corresponding author: Emily G. Hurley, MD University of Cincinnati College of Medicine Department of Obstetrics and Gynecology 231 Albert Sabin Way, Mail Location 0526 Cincinnati, Ohio 45267-0526 United States of America Cell phone: 1-513-919-7696 Fax: 1-513-558-3558 Email: [email protected]
32
AC C
Word count: Abstract 333 Main text 2,117
EP
TE D
M AN U
SC
8 9 10
Table 3 is to appear in the print issue should the submission be accepted for publication.
ACCEPTED MANUSCRIPT 2 33
Condensation: Older paternal age does not increase risk of adverse perinatal outcomes.
34
Short version of article title: PATERNAL AGE AND PERINATAL OUTCOMES
RI PT
35 36
AC C
EP
TE D
M AN U
SC
37
ACCEPTED MANUSCRIPT 3
Abstract
39
Background: There is an increasing trend to delay childbearing to advanced parental age.
40
Increased risks of advanced maternal age and assisted reproductive technologies are widely
41
accepted. There is limited data regarding advanced paternal age. In order to adequately
42
counsel patients on risk, more research regarding advanced paternal age is necessary.
43
Objective: To determine the influence of paternal age on perinatal outcomes, and to assess
44
whether this influence differs between pregnancies achieved spontaneously and those
45
achieved with assisted reproductive technology (ART).
46
Study Design: A population-based retrospective cohort study of all live births in Ohio from
47
2006-2012 was completed. Data were evaluated to determine if advanced paternal age is
48
associated with an increased risk of adverse outcomes in pregnancies. The analysis was
49
stratified by status of utilization of ART. Generalized linear regression models assessed the
50
association of paternal age on pregnancy complications in ART and spontaneously conceived
51
pregnancies, after adjusting for maternal age, race, multifetal gestation and Medicaid status,
52
using STATA software.
53
Results: Paternal age was documented in 82.2% of 1,034,552 live births in Ohio during the 7-
54
year study period. Paternal age ranged from 12-87 years, median 30 (IQR 26, 35). Maternal
55
age ranged from 11-62 years, median 27 (IQR 22, 31). The use of ART in live births increased as
56
paternal age increased: 0.1% <30 years vs. 2.5% > 60 years, p<0.001. After accounting for
57
maternal age and other confounding risk factors, increased paternal age was not associated
58
with a significant increase in the rate of pre-eclampsia, preterm birth, fetal growth restriction,
59
congenital anomaly, genetic disorder or Neonatal Intensive Care Unit admission. The influence
AC C
EP
TE D
M AN U
SC
RI PT
38
ACCEPTED MANUSCRIPT 4
of paternal age on pregnancy outcomes was similar in pregnancies achieved with ART
61
compared to non-ART pregnancies.
62
Conclusion: Older paternal age does not appear to pose an independent risk of adverse
63
perinatal outcomes, either in pregnancies achieved with or without ART. However, small effect
64
sizes such as very small risk increases or decreases may not be detectable despite the large
65
sample size in this study of over 830,000 births.
SC
66
Keywords: Advanced paternal age; assisted reproductive technology; perinatal outcomes
M AN U
67
RI PT
60
68
AC C
EP
TE D
69
ACCEPTED MANUSCRIPT 5 70 71
Introduction There is an increasing trend to delay childbearing to advanced parental age, mother and father alike. The associated risks of advanced maternal age, generally defined ≥35 years of
73
age, are well studied. Increased maternal age is associated with increased miscarriage rate,
74
obstetric complications and perinatal mortality and is also associated with decreased success
75
when using assisted reproductive technologies (ART).1-2 Less is known regarding the risks
76
associated with advanced paternal age and pregnancy outcomes. Some studies have shown
77
that advanced paternal age has effects on sperm quality, miscarriage rates, infant birth weights,
78
and childhood/adult illnesses.1,3-8 Nonetheless, one study published in 2008 by Chen et al.
79
found that advanced paternal age is not associated with adverse birth outcomes.9
SC
M AN U
80
RI PT
72
As parental ages increase, more couples may turn to ART for aid in achieving pregnancy. ART itself is associated with increased perinatal risks including prematurity, low birth weight
82
and multiple gestations.10 However, there is a paucity of data evaluating the impact of
83
advanced paternal age on the risks associated with ART pregnancies. A review of 10 articles
84
published in 2011 found that there is insufficient evidence available to determine if paternal
85
age has a negative impact on ART outcomes. The outcomes evaluated included fertilization
86
rate, pregnancy rate, miscarriage rate and live birth rate. Only four of the studies reviewed
87
evaluated live birth rate with two of the studies finding no significant difference and two
88
studies with a decreased chance of live birth with advanced paternal age.1 A study published in
89
2014 found no increased risk for prematurity, low birthweight or small for gestational age in
90
pregnancies achieved with ART at older paternal age.11 After a thorough review of the
AC C
EP
TE D
81
ACCEPTED MANUSCRIPT 6 91
literature, overall there is a limited amount of data regarding advanced paternal age and
92
perinatal outcomes in pregnancies achieved with ART. The aim of this study was to utilize a large population-based data source to assess
RI PT
93
whether advanced paternal age poses independent increased perinatal risks, and assess
95
whether those risks differ in pregnancies achieved by ART or those achieved spontaneously
96
without ART.
97
Material and Methods
99
A population-based retrospective cohort study of all live births in Ohio from 2006-2012
M AN U
98
SC
94
was performed to quantify the influence of advanced paternal age on pregnancy complications. Study approval was obtained from the Ohio Department of Health Institutional Review Board
101
and a de-identified data set extracted from live-birth certificates was provided for this analysis.
102
This study was exempt from review by the Institutional Review Board at the University of
103
Cincinnati, Cincinnati, Ohio.
104
TE D
100
The exposure of interest was paternal age. Paternal age was evaluated both continuously and categorically (<30, 30-39, 40-49, 50-59 and >60 years of age). The 2003
106
version of the birth certificate includes three areas to indicate use of fertility treatment under
107
the header “RISK FACTORS IN THIS PREGNANCY.” The initial option included is pregnancy
108
resulting from infertility treatment, and if chosen, the two further options include 1.) use of
109
fertility-enhancing drugs, artificial insemination or intrauterine insemination and 2.) assisted
110
reproductive technology (in vitro fertilization, gamete intrafallopian transfer).12 Guidelines and
111
instructions for identification of use of infertility treatments resulting in live births in the US is
112
outlined in the National Center for Health Statistics Guide to Completing the Facility
AC C
EP
105
ACCEPTED MANUSCRIPT 7
Worksheets for the Certificate of Live Birth and Report of Fetal Death (2003 revision).13 For this
114
study, ART status was assigned as “assisted reproductive technology” recorded on page 2, in
115
field number 41 on the live birth certificate.12
116
RI PT
113
Outcomes of interest for this study included: pre-eclampsia, preterm birth, fetal growth restriction, congenital anomaly, genetic disorder, and Neonatal Intensive Care Unit (NICU)
118
admission. Pre-eclampsia included cases recorded on the birth certificate as gestational
119
hypertension, pregnancy induced hypertension (PIH), or pre-eclampsia. Preterm birth was
120
defined as gestational age at time of birth recorded as <37 weeks by obstetric estimate, which
121
includes data from both last menstrual period and ultrasound estimates of gestational age.
122
Fetal growth restriction was defined as birthweight <10th percentile using a large US national
123
growth reference.14 Any major congenital anomaly included newborns with anencephaly,
124
meningomyelocele/spina bifida, congenital heart disease, congenital diaphragmatic hernia,
125
omphalocele, gastroschisis, limb reduction defect, cleft lip and/or cleft palate, or hypospadias
126
indicated in field number 55 of the birth certificate. Genetic disorder included cases of Down
127
syndrome or other suspected chromosomal disorder at the time of birth certificate recording.
128
NICU admission included newborns admitted to the Neonatal Intensive Care Unit during the
129
first few days of life, again as recorded at the time of birth certificate generation.12
M AN U
TE D
EP
AC C
130
SC
117
To quantify the effect of paternal age on the dichotomous outcomes of adverse
131
perinatal outcomes including pre-eclampsia, preterm birth, fetal growth restriction, congenital
132
anomaly, genetic disorder, and NICU admission, we used a log-binomial model, Generalized
133
Linear Model (GLM), with log link. This linear model approach for estimating relative risk, is
134
well-suited for dichotomous outcomes. Risk estimates for pregnancy complications were
ACCEPTED MANUSCRIPT 8
stratified by ART and non-ART pregnancies. Adjusted relative risk (RR) values and 95%
136
confidence intervals were calculated to estimate the risk of advancing paternal age on each
137
adverse pregnancy outcome after adjustment for the confounding influences of advancing
138
maternal age, race, Medicaid status, and multifetal gestation. Sensitivity analyses were
139
performed estimating relative risk with maternal age held constant, and also imputing missing
140
data on paternal age as a spectrum of ages. Statistical analysis was performed using STATA
141
Release 12 Software (StataCorp, College Station, TX, USA).
142
Results
SC
M AN U
143
RI PT
135
Paternal age was documented in approximately 82.2% of 1,034,552 live births in Ohio between the years of 2006-2012. A total of 833,727 births were included in the study after
145
excluding those without a paternal age recorded and limiting to the first birth of multifetal
146
gestations. Approximately 3.7% of the births (3,118) were achieved with ART compared to
147
830,609 non-ART births. Paternal age ranged from 12-87 years, median 30 (IQR 26, 35).
148
Maternal age ranged from 11-62 years, median 27 (IQR 22, 31). As paternal age increased, the
149
corresponding maternal age also increased, however not linearly. When mothers were <20
150
years of age, the majority (96.2%) of fathers were <30 years of age. When the paternal age was
151
>40, the majority of mothers were >35 years of age (Table 1).
EP
AC C
152
TE D
144
Parental demographic characteristics of the live birth population are displayed in Table
153
1. There were higher rates of both mothers and fathers with less than a high school education
154
when the paternal age was in the youngest (<30 years) and oldest (>60) categories. Older
155
paternal age was associated with a higher proportional frequency of non-Hispanic black
ACCEPTED MANUSCRIPT 9 156
maternal race. Utilization of state-funded Medicaid insurance and maternal tobacco use were
157
higher in pregnancies with young and old paternal age. The frequency of perinatal complications was higher in the youngest and oldest paternal
RI PT
158
age groups, as displayed in Table 2. In unadjusted analysis, prior to accounting for the
160
confounding risk of maternal age, the crude risk of pregnancy complications was highest in the
161
extreme categories of paternal age. The paternal age category with the lowest frequency and
162
unadjusted risk of pregnancy complications was 30-39 years of age (Table 2). However,
163
adjustment for maternal age resulted in null independent influence of paternal age on adverse
164
pregnancy outcomes, indicating that the observed influence of paternal age was attributable
165
primarily to concordant similar extremes of maternal age (Table 3). These findings were similar
166
in pregnancies utilizing ART and those conceived without the assistance of ART. Sensitivity
167
analysis performed removing maternal age from the model and holding maternal age constant,
168
and also imputing missing data on paternal age as a variety of ages did not significantly alter the
169
relative risk effect estimates displayed in Table 3.
M AN U
TE D
The use of ART in live births significantly increased as paternal age increased with 0.1%
EP
170
SC
159
of pregnancies achieved with ART when paternal age was <30 compared to 2.5% when >60
172
years of age (p<0.001) (Table 4). There was a significant interaction between the oldest
173
paternal age category, ≥60 years and ART (p=0.02), and stratified analyses by ART status were
174
performed. After accounting for the confounding influences of maternal age, race, Medicaid
175
status and multifetal gestation, increased paternal age was not associated with a significant
176
increase in the rate of pre-eclampsia, preterm birth, fetal growth restriction, congenital
AC C
171
ACCEPTED MANUSCRIPT 10
anomaly, genetic disorder or NICU admission with analyses stratified by use of ART, with all
178
adjusted relative risk estimate 95% confidence intervals crossing the null value of 1 (Table 3).
179
Comment
RI PT
177
As couples delay childbearing, the question of the effects of older age gain importance.
181
Advanced maternal age is well defined with understood risks and implications, however, this is
182
not true for advanced paternal age. There is no widely accepted definition of advanced
183
paternal age and the consequences of pregnancies achieved at advanced paternal age are not
184
well understood. Also, with advanced parental age, couples are more likely to utilize ART which
185
has risks of its own. Our study demonstrates that advanced paternal age does not increase the
186
risk of poor perinatal outcomes, even when stratified by pregnancies achieved with ART
187
compared to spontaneous pregnancies. After adjusting for maternal age and other
188
confounding risk factors, the risk of pre-eclampsia, preterm birth, fetal growth restriction,
189
congenital anomaly, genetic disorder or NICU admission was not increased with advancing
190
paternal age.
EP
TE D
M AN U
SC
180
Our study is consistent with a previous study completed by Stern et al., which was
192
published in 2014. The study was smaller (n=280,804) and only evaluated the effect of paternal
193
age on three outcomes including prematurity, low birthweight and small for gestational age.
194
The outcomes were limited to mothers aged 40 and younger. Nonetheless, the study found
195
that older paternal age was not associated with increased risk for prematurity, low birthweight
196
or small for gestational age in pregnancies in fertile, subfertile and infertile populations using
197
ART.11
AC C
191
ACCEPTED MANUSCRIPT 11
A previous population-based retrospective cohort study published by Yang et al. found
199
that older paternal age is associated with an increased risk of birth defects. The study by Yang
200
and colleagues had a much larger sample size with over 5 million subjects, a sample size larger
201
than six times the 833,727 births included in our study. 15 Our study also identified an increased
202
relative risk of congenital anomalies with increasing paternal age similar to the effect size
203
identified in the study by Yang (Table 3), however it was not statistically significant. The
204
absolute risk increase observed by Yang was only 0.4% increased prevalence, therefore the
205
significance of the result may not be clinically meaningful and may represent alpha error. The
206
difference between the two studies may be a reflection of the difference in the sample sizes
207
however is not a difference in effect size.
M AN U
SC
RI PT
198
Limitations to this study include the retrospective use of birth certificate data, both in
209
regards to paternal age and use of ART. Although there are published guidelines and training
210
on how to fill out the US birth certificate, the accuracy of recorded variables may vary.
211
Approximately 17.8% of births recorded in Ohio from 2006-2012 did not have paternal age
212
recorded, which may reflect unknown paternity. Also we were unable to assess whether the
213
recorded paternal age took into consideration whether a sperm donor was used. A study
214
published in July 2016 compared the Society for Assisted Reproductive Technology Clinic Online
215
Reporting System to birth certificate data of 8 states and found that only 36.5% of children
216
conceived utilizing in vitro fertilization was accurately reported on the birth certificate.16 We
217
would presume that those recorded on the birth certificate as ART were accurate, however
218
there may be a significant number of ART cases listed as no ART, or missing data for that field.
219
Thus, our results may underestimate the risks in ART pregnancies as some ART pregnancies may
AC C
EP
TE D
208
ACCEPTED MANUSCRIPT 12
be included in the referent group. However, we presume this influence would be minimal as no
221
significant risk increase for outcomes was observed overall, or in the stratified analysis among
222
ART and non-ART pregnancies. Furthermore, sensitivity analyses modeling missing data on
223
paternal age as a variety of young, mid-range, and older paternal ages did not significantly alter
224
our observed relative risk estimates.
Another limitation of this study is that long-term outcomes cannot be evaluated. Given
SC
225
RI PT
220
that our findings are based on birth certificate data, we were unable to assess later diagnosed
227
childhood or adult illness of infants born to fathers of older age. Also, this study is limited to
228
the state of Ohio, which has a population that may not be generalizable to all pregnancies
229
across the United States and in other countries. In addition, as in any observational study,
230
unidentified and unmeasured confounds exist which are unable to be accounted for. A major
231
strength of this study was the large sample size.
TE D
M AN U
226
In summary, we found that advanced paternal age is not significantly associated with an
233
independent increase in risk of adverse perinatal outcomes. The observed increased frequency
234
of adverse outcomes at the extremes of paternal age became non-significant after adjusting for
235
the confounding influence of maternal age and other risk factors. These findings suggest that
236
maternal age is the primary contributor to adverse outcomes in couples who delay childbearing
237
to older parental ages. This information may be clinically useful when counseling couples
238
planning on parental age-related pregnancy risks.
239 240
AC C
EP
232
ACCEPTED MANUSCRIPT 13
Acknowledgements
242
This study includes data provided by the Ohio Department of Health, which should not be
243
considered an endorsement of this study or its conclusions.
RI PT
241
244 245
SC
246
M AN U
247 248
AC C
EP
TE D
249
ACCEPTED MANUSCRIPT 14
251
References 1.
outcome. Fertil Steril. 2011;95(1):1–8.
252 253
Dain L, Auslander R, Dirnfeld M. The effect of paternal age on assisted reproduction
2.
RI PT
250
Valent AM, Newman T, Chen A, Thompson A, Defranco E. Gestational age-specific neonatal morbidity among pregnancies complicated by advanced maternal age: A
255
population-based retrospective cohort study. J Matern Neonatal Med.
256
2016;29(9):1485–90. 3.
Sharma R, Agarwal A, Rohra V, Assidi M, Abu-Elmagd M, Turki RF. Effects of
M AN U
257
SC
254
258
increased paternal age on sperm quality, reproductive outcome and associated
259
epigenetic risks to offspring. Reprod Biol Endocrinol. 2015;13(1):35.
260
4.
de la Rochebrochard E, Thonneau P. Paternal age and maternal age are risk factors for miscarriage; results of a multicentre European study. Hum Reprod.
262
2002;17(6):1649–56. 5.
6.
7.
270
Reichman NE, Teitler JO. Paternal age as a risk factor for low birthweight. Am J Public
Health. 2006;96(5):862–6.
268 269
Slama R, Bouyer J, Windham G, Fenster L, Werwatz A, Swan SH. Influence of paternal age on the risk of spontaneous abortion. Am J Epidemiol. 2005;161(9):816–23.
266 267
EP
and spontaneous abortion. Obstet Gynecol. 2006;108(2):369–77.
264 265
Kleinhaus K, Perrin M, Friedlander Y, Paltiel O, Malaspina D, Harlap S. Paternal age
AC C
263
TE D
261
8.
Ramasamy R, Chiba K, Butler P, Lamb DJ. Male biological clock: a critical analysis of advanced paternal age. Fertil Steril. 2015;103(6):1402–6.
ACCEPTED MANUSCRIPT 15
9.
Chen X, Wen S, Krewski D, Fleming N, Yang Q, Walker M. Paternal age and adverse birth outcomes: Teenager or 40+, who is at risk? Hum Reprod. 2008;23(6):1290–6.
272 273
10.
reproductive technology. Obstet Gynecol. 2005;106(5 Pt 1):1143–6.
274 275
ACOG Committee Opinion #324: Perinatal risks associated with assisted
RI PT
271
11.
Stern JE, Luke B, Hornstein MD, et al. The effect of father’s age in fertile, subfertile, and assisted reproductive technology pregnancies: A population based cohort study.
277
J Assist Reprod Genet. 2014;31(11):1437–44. 12.
http://www.cdc.gov/nchs/data/dvs/birth11-03final-ACC.pdf.
279 280
U.S. Standard Certificate of Live Birth. Revised 11/2003. Available from:
M AN U
278
SC
276
13.
National Center for Health Statistics. Guide to Completing the Facility Worksheets for the Certificate of Live Birth and Report of Fetal Death (2003 revision). Hyattsville,
282
MD. US Department of Health and Human Services, Centers for Disease Control and
283
Prevention, 2016. Available from:
284
http://www.cdc.gov/nchs/data/dvs/GuidetoCompleteFacilityWks.pdf. 14.
reference for fetal growth. Vol. 87, Obstetrics and gynecology. 1996. p. 163–8.
286
288 289
15.
Yang Q, Wen SW, Leader A, Chen XK, Lipson J, Walker M. Paternal age and birth
AC C
287
Alexander GR, Himes JH, Kaufman RB, Mor J, Kogan M. A United States national
EP
285
TE D
281
defects: how strong is the association? Human Reproduction. 2007; 22: 696-701.
16.
Luke BB, Brown MB, Spector LG. Validation of infertility treatment and assisted
290
reproductive technology use on the birth certificate in eight states. Vol. 215. Am J
291
Obstet Gynecol. 2016. p. 126-127.
292
ACCEPTED MANUSCRIPT 16 293
Table 1. Baseline characteristics stratified by paternal age categories.
Characteristics
Paternal Age
p-value
30-39
40-49
50-59
≥ 60
N=373,649
N=379,942
N=73,075
N=6,459
N=602
(36.8%)
(37.4%)
(7.2%)
(0.6%)
(0.06%)
<20
56,844 (15.2%)
1,954 (0.5%)
223 (0.3%)
35 (0.5%)
3 (0.5%)
20-34
311,982(83.5%) 314,813 (82.9%)
31,188 (42.7%)
2,737 (42.1%)
255 (42.4%)
>35
3,408 (0.9%)
41,420 (56.7%)
3,655 (56.6%)
342 (56.8%)
M AN U
62,010 (16.3%)
SC
Maternal age
RI PT
<30
White
301,707(80.7%) 320,893 (84.5%)
58,093 (79.5%)
4,465 (69.1%)
353 (58.6%)
Black
46,055 (12.3%)
29,307 (7.7%)
8,766 (12%)
1,273 (19.7%)
158 (26.2%)
Hispanic
20,554 (5.5%)
13,861 (3.6%)
2,609 (3.6%)
281 (4.3%)
20 (3.3%)
Other
5,333 (1.4%)
TE D
Maternal Race/Ethnicity
15,881 (4.2%)
3,607 (4.9%)
440 (6.8%)
71 (11.8%)
<0.001
<0.001
58,791 (15.7%)
21,659 (5.7%)
4,333 (5.9%)
507 (7.8%)
66 (11.0%)
<0.001
Maternal
71,650 (19.2%)
28,332 (7.5%)
6,554 (9.0%)
860 (13.3%)
114 (18.9%)
<0.001
Maternal Tobacco Use
103,541(27.7%) 60,822 (16%)
11,699 (16%)
1,408 (21.8%)
122 (20.3%)
<0.001
Maternal Medicaid use
164,045(42.9%) 69,287 (18.2%)
14,556 (19.9%)
1,945 (30.1%)
202 (33.5%)
<0.001
Mother Unmarried
178,702(47.8%) 62,979 (16.6%)
12,374 (16.9%)
1,554 (24.1%)
139 (23.1%)
<0.001
AC C
294 295 296 297
EP
Paternal
ACCEPTED MANUSCRIPT 17 Table 2. Pregnancy complications and delivery characteristics stratified by paternal age.
Maternal-fetal complication
Fetal growth restriction RR (95% CI) Major fetal anomaly RR (95% CI) Genetic disorder RR (95% CI)
299 300
≥ 60
Missing age
N=373,649
N=379,942
N=73,075
N=6,459
N=602
N=180,841
(36.8%)
(37.4%)
(7.2%)
(0.6%)
(0.06%)
19,337 (5.2%)
17,970 (4.8%)
3,814 (5.3%)
362 (5.7%)
32 (5.4%)
9701 (5.5%)
1.06 (1.04,1.08)
0.92 (0.90,0.94)
1.07 (1.04,1.10)
1.15 (1.04,1.27)
1.12 (0.82,1.54)
1.09 (1.07,1.12)
40,495 (10.8%)
38,284 (10.1%)
8,999 (12.3%)
959 (14.8%)
86 (14.3%)
28,838 (16.6%)
1.00 (0.99,1.01)
0.92 (0.92,0.94)
1.19 (1.17,1.21)
1.40 (1.33,1.48)
1.38 (1.16,1.65)
1.44 (1.43,1.46)
37,370 (10%)
29,750 (7.8%)
6,390 (8.7%)
704 (10.9%)
72 (12.0%)
26,913 (15.5%)
1.22 (1.20,1.24)
0.81 (0.80,0.82)
1.00 (0.98,1.02)
1.23 (1.15,1.31)
1.35 (1.09,1.65)
1.65 (1.63,1.67)
1,316 (0.35%)
1,085 (0.29%)
238 (0.33%)
17 (0.26%)
2 (0.33%)
691 (0.40%)
1.20 (1.11,1.29)
0.82 (0.76,0.89)
1.05 (0.92,1.20)
0.85 (0.53,1.34)
0.99 (0.25,3.93)
1.22 (1.12,1.32)
242 (0.06%)
373 (0.10%)
178 (0.24%)
24 (0.37%)
3 (0.50%)
184 (0.10%)
0.52 (0.44,0.60)
1.00 (0.87,1.15)
2.84 (2.41,3.35)
3.76 (2.51,5.65)
4.84 (1.56,15.0)
1.04 (0.89,1.23)
RI PT
50-59
M AN U
RR (95% CI)
40-49
TE D
Preterm birth <37 weeks
30-39
EP
RR (95% CI)
p-value
<30
AC C
Pre-eclampsia
Paternal Age
SC
298
<0.001
<0.001
<0.001
<0.001
<0.001
ACCEPTED MANUSCRIPT 18 Table 3. Influence of increasing paternal age on the risk of adverse pregnancy outcomes in ART and non-ART pregnancies. Non-ART Pregnancies
N=3,118
N=830,609
adjRR (95% CI)*
Crude RR
adjRR (95% CI)*
Pre-eclampsia
1.01 (0.99, 1.03)
1.01 (0.98, 1.03)
1.01 (1.00, 1.01)
1.00 (0.99 ,1.00)
Preterm birth <37 weeks
1.00 (1.02, 1.02)
1.00 (0.99, 1.01)
1.02 (1.02, 1.02)
1.01 (1.01, 1.01)
Fetal growth restriction 0.99 (0.97, 1.01) (birthweight <10th percentile)
1.00 (0.98, 1.03)
Congenital anomaly (any)
1.00 (0.94, 1.07)
0.99 (0.88, 1.11)
Genetic disorder
1.08 (0.94, 1.24)
NICU admission
1.00 (0.98, 1.01)
1.01 (1.01, 1.01)
1.01 (1.01, 1.02)
1.01 (1.00, 1.02)
1.01 (0.99, 1.03)
1.06 (0.88, 1.27)
1.09 (1.07, 1.10)
1.01 (0.99, 1.04)
1.00 (0.98, 1.02)
1.02 (1.02, 1.02)
1.01 (1.01, 1.01)
EP
TE D
*Relative risk limited to paternal age ≥ 30 years, adjusted for maternal age, multifetal birth, maternal race, and Medicaid status, (95% confidence interval).
AC C
303 304 305 306 307 308
SC
Crude RR
M AN U
Adverse Pregnancy Outcome
ART Pregnancies
RI PT
301 302
ACCEPTED MANUSCRIPT 19 Table 4. Use of infertility treatment stratified by paternal age categories. Infertility Treatment Paternal Age 40-49
50-59
≥ 60
N=373,649
N=379,942
N=73,075
N=6,459
N=602
(36.8%)
(37.4%)
(7.2%)
(0.6%)
(0.06%)
1,271
5,210
1,533
151
22
(0.34%)
(1.38%)
(2.11%)
(2.36%)
(3.67%)
428
1,912
667
(0.12%)
(0.51%)
(0.92%)
96
15
(1.5%)
(2.5%)
M AN U
Assisted reproductive technologies
30-39
RI PT
Infertility Drugs and/or insemination
p-value
<30
SC
309
AC C
EP
TE D
310
<0.001
<0.001