If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Pediatrics, Mount Sinai Hospital, and the Departments of Pediatrics and of Health Policy, Management and Evaluations, University of Toronto, Toronto, Ontario, Canada
A systematic review of the risks of a low birthweight (LBW), preterm, and small-for-gestational-age births in relation to paternal factors was performed. Medline, Embase, Cumulative Index of Nursing and Allied Health Literature, and bibliographies of identified articles were searched for English-language studies. Study qualities were assessed according to a predefined checklist. Thirty-six studies of low-to-moderate risk of bias were reviewed for various paternal factors: age, height, weight, birthweight, occupation, education, and alcohol use. Extreme paternal age was associated with higher risk for LBW. Among infants who were born to tall fathers, birthweight was approximately 125-150 g higher compared with infants who were born to short fathers. Paternal LBW was associated with lower birthweight of the offspring. In conclusion, paternal characteristics including age, height, and birthweight are associated with LBW. Paternal occupational exposure and low levels of education may be associated with LBW; however, further studies are needed.
Low birthweight (LBW) birth and preterm births (PTBs) are public health issues with significant individual, familial, and societal impact. Familial influence in the cause of LBW/PTB births is suspected; however, the major research attention has been focused on maternal determinants. Paternal factors (such as advanced paternal age
in a minireview, concluded that advanced paternal age (>40 years) was associated with miscarriages and fetal deaths. Diseases such as Alpert syndrome, Marfan syndrome, and Waardenberg syndrome are associated with advanced paternal age. Strobino et al
reviewed the connection between paternal occupational exposure and effect on offspring. They reported that paternal occupation was not associated with spontaneous abortion and that there was a lack of conclusive impact on other birth outcomes.
Studies have reported paternal factors that influence birthweight or gestational age. However, no comprehensive review of paternal determinants or factors on LBW, PTB, or small for gestational age (SGA) births has been conducted. The objective of this study was to review systematically the risk of an infant with LBW, PTB, and SGA in relation to various reported paternal factors.
Materials and Methods
The data were extracted from published articles; therefore, no ethical approval was obtained.
Criteria for consideration of studies for this review
Observational studies that explored the association of any of the paternal factors and the outcomes of LBW, PTB, and SGA births of offspring were included in this review. If the study provided adequate information on the method of ascertainment of the paternal factor and its effects on any of the outcomes of interest, the study was eligible for inclusion in the review. We included only information that was available from the publications and did not contact primary authors. Studies that were published as abstracts were excluded.
Types of studies
Observational cohort studies, case control studies, and studies of surveys or interviews were included. Reports of data from national or local vital statistics that were not published as peer-reviewed articles were not included.
Types of participants
Women who had a live birth were included. Data on reports of maternal influence on birth outcomes were not included in this review.
Assessment of exposure
Previous knowledge of the subject indicated that paternal age, anthropometry, paternal birthweight, occupation, and educational background would be probable exposures to be included in the review.
We did not include paternal race as an exposure variable because race has a complex interaction with maternal race. Similarly, we excluded paternal smoking as an exposure because most studies have reported on environmental tobacco exposure, which could have been from partners or other sources, and it was difficult to differentiate. We included studies that reported a collection of ascertainment of exposure data from maternal charts, interview, and direct or indirect assessments.
Types of outcome measures
Studies that reported data on any of the following outcomes were included: (1) LBW birth defined as birthweight <2.5 kg; (2) PTB defined as gestational age <37 weeks; (3) birthweight in grams; (4) gestational age in weeks, and (5) SGA birth defined as birthweight <10th percentile for gestational age.
Search strategy for identification of studies
Electronic databases (Medline, Embase, and Cumulative Index of Nursing and Allied Health Literature) were searched, with assistance from an experienced librarian, from their inception to March 2009 for all published studies in the English language. The search terms were modified according to database requirements. The reference lists of the identified articles were reviewed to locate additional eligible studies. The articles were scanned initially on the basis of titles and abstracts. The reviewer was not blinded to authors or institution. Selected articles were retrieved in full and were assessed for eligibility. Search terms used were low birthweight; premature birth; preterm birth; small for gestational age; growth, intrauterine; growth restriction, fetal; growth restriction, intrauterine; high risk pregnancy; infant, premature; infant, newborn; pregnancy; familial factors; father, paternal factors, paternal age, paternal birthweight, occupation, education, anthropometry, weight, height, body mass, and body mass index.
Methods of the review
Data extraction
Data from each eligible study were extracted into custom-made data collection forms. Minor modifications (such as combining birthweight means for male and female subjects) and the calculation of raw percentages from available data were done. Confounders that were adjusted for in the analyses in the individual studies were reported. When adjusted data were reported in the primary studies, they were extracted and noted in the results.
Assessment of quality of included studies
The methodologic quality of studies was assessed with a predefined checklist that was based on criteria for the sample selection, exposure assessment, outcome assessment, confounder, and analytical and attrition biases (Appendix). The classifications were applied in each category: cannot tell, no bias, low risk, moderate risk, or high risk of bias.
Data synthesis
We expected a significant degree of clinical heterogeneity among studies in each of the paternal factors and planned for a systematic review and not metaanalyses from the outset. Qualitative synthesis of the evidence was planned because of the reasons mentioned earlier.
Heterogeneity and publication bias assessment
Clinical heterogeneity was assessed and reported in the Tables.
Results
Description of studies
Thirty-eight studies were included in this review.
Paternal occupational exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin and birth outcomes of offspring: birth weight, preterm delivery, and birth defects.
Effect of paternal alcohol consumption before conception on infant birth weight: ALSPAC Study Team: Avon Longitudinal Study of Pregnancy and Childhood.
Some of the included studies reported on >1 paternal factor. Paternal alcohol use was identified as an exposure during the literature review (which was not planned a priori). Eleven studies were excluded after detailed evaluation. Details of included studies and reasons for excluded studies are given in Figure 1.
The results of the quality assessments of the included studies are reported in TABLE 1, TABLE 2, TABLE 3, TABLE 4, TABLE 5, TABLE 6. The studies had low-to-moderate risk of biases. The major criticism of the included studies was that the assessment of exposure in many reports was indirect (such as maternal history, derived or deduced assessment method for occupational exposure). Adjustment of confounders was variable among studies; however, compared with other reports of determinants of LBW/PTB births, most of the studies in this category have accounted for common confounders. Assessment of outcomes was satisfactory in most studies. Many studies reported only incidence or percentage of outcome data and not odds ratio or relative risk and their associated confidence interval. The data in the Tables reflect what is reported in the articles. If the risk estimates are missing from a Table, the implication is that they were not reported.
TABLE 1Paternal age and relation to low birthweight, preterm, and small-for-gestational-age births
Paternal weight and BMI were not significant predictor of birthweight once maternal weight (P = .16) and BMI (P = .42) were taken in account, respectively
Paternal weight divided in 9 strata (5 kg strata for <50 to >90 kg) was not correlated with birthweight (P = .052); paternal BMI divided in 4 strata (5 kg/m2 strata for <20 to >35 kg/m2) was not correlated with birthweight (P = .33)
1959-1961; Cohort study, fathers born in this period derived from Danish population register and their singleton, liveborn, children born from 1974-1989; n = 1258
2001; Comparative cohort study; fathers who worked as painters (472 cases) were compared with carpenters (462 control subjects) in The Netherlands; n = 934
Organic solvents for painters assessed by questionnaire
Outcome: questionnaire
Selection: low
Low level (0.17–0.38)
1.5 (0.5–4.3)
1.6 (0.7–3.9)
Confounders adjusted: maternal age, smoking, alcohol use, chemical exposure; paternal smoking, alcohol use
Paternal occupational exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin and birth outcomes of offspring: birth weight, preterm delivery, and birth defects.
1981-1992; Matched cohort study; fathers exposed to high lead levels (>40 μg/dL before 1986 and >25 μg/dL after 1986 (n = 747) compared with male bus drivers (n = 2259) in New York, 1974-1989 (n = 1258)
Lead exposure assessed from laboratory reports of workers
Outcome: birth certificate
Selection: low
General exposure (risk ratio)
1.00 (0.67–1.50)
0.89 (0.64–1.26)
0.86 (0.64–1.15)
Confounders adjusted: paternal age, maternal education, maternal complications, prenatal care, race, parity, infant sex
1961-1971; Matched cohort study; children of fathers exposed to herbicide spray during Vietnam conflict (n = 859) compared with children of unexposed Air Force veterans (n = 1223); n = 2082
2,3,7,8-Tetrachloro-dibenzo-p-dioxin assessed from blood sample and modeled calculation
1981-1989; Case control study; control subjects enrolled in another study in Baltimore-Washington DC; cases (n = 220) were LBW; controls were not LBW (n = 522); n = 742
Lead exposure assessed by history regarding type of job and hours exposed
Effect of paternal alcohol consumption before conception on infant birth weight: ALSPAC Study Team: Avon Longitudinal Study of Pregnancy and Childhood.
There was variability in age cut-offs that were used in the studies as well as the reference age for comparison of groups. Exploratory analysis of a relationship between paternal age and LBW births was conducted. Figure 2 demonstrates the relationship of paternal age (mid values from all reported age groups and conversion of the 25- to 28-year age group as the reference group for each study) and LBW births in 6 studies that reported the odds of LBW births in various age groups. A “saucer-shaped” association can be inferred with higher odds at both extremes. No association between paternal age and PTB or SGA births was identified in any of the studies. Characteristics of the included studies, assessment of biases, and reported results are outlined in Table 1.
FIGURE 2Scatter plot of paternal age and odds ratio for low birthweight (LBW) births that were reported in various age groups.
Paternal height showed significant correlation with birthweight of the offspring in most of the studies. On average, there was 125-150 g reduction in birthweight of the offspring of short vs tall fathers. Characteristics of the included studies, assessment of biases, and reported results are given in Table 2.
Paternal weight and body mass
Most of the studies reported a nonsignificant relationship with paternal body weight or mass and birthweight and LBW births. In 1 study, a significant relationship became nonsignificant when maternal body mass was considered. Characteristics of the included studies, assessment of biases, and reported results are given in Table 3.
Paternal birthweight
Three studies reported that an infant's birthweight increased as paternal birthweight increased. The odds of a LBW birth were higher if the father had a LBW birth. Interestingly, 1 study reported higher rates of PTBs as the father's birthweight increased; however, this might be attributed to maternal characteristic. Characteristics of the included studies, assessment of biases, and reported results are given in Table 4.
Paternal occupation
Higher and prolonged lead exposures were associated with higher risk of LBW birth and PTB; however, details of timing of exposure in relation to pregnancy were not available in detail to assess direct effect. One large study reported a higher adjusted risk of SGA births among workers who were exposed to benzene, chromium, and other minerals and a higher risk of PTBs among workers who were exposed to x-rays. Other occupations that were assessed were not associated with higher risk of adverse outcomes. Characteristics of the included studies, assessment of biases, and reported results are given in Table 5.
Paternal education
One study reported a higher risk of LBW births and another reported a higher risk of PTBs among fathers who had a high school education only, compared with fathers who had a college education. Characteristics of the included studies, assessment of biases, and reported results are given in Table 6.
Paternal alcohol use
One small study reported a 200-g reduction in birthweight with regular alcohol use; other studies reported no difference in the risk of LBW birth or PTBs with different levels of paternal alcohol use.
Effect of paternal alcohol consumption before conception on infant birth weight: ALSPAC Study Team: Avon Longitudinal Study of Pregnancy and Childhood.
Characteristics of the included studies, assessment of biases, and reported results are given in Table 6.
Comment
In this systematic review of 36 studies on paternal factors and birth outcomes, certain positive and negative associations were observed. Extreme paternal ages (<20 and >40 years) may be associated with LBW; however, no consistent associations with PTBs and SGA births were identified. Paternal height and paternal occupational exposure to lead were associated with LBW and birthweight of the offspring. The infant's birthweight increased with a higher paternal birthweight. Paternal weight, body mass, and occupational exposures to herbicides, plant work, woodwork and paternal alcohol use were not associated with birth outcomes. A lower level of paternal education was associated with LBW/PTB and SGA births; however, confirmation of data from larger studies is warranted.
Paternal age
Advanced paternal age has been linked with fetal loss,
Biologic rationale behind the influence of paternal age on birth outcomes stems from the identification of greater expression of paternal genes on the placenta
The overall impression from our systematic review of included studies and the plot of paternal age–LBW birth incidence curve suggest that, if there was an effect of paternal age on LBW birth, it would appear to be modest for advanced paternal age.
Paternal anthropometry
All 10 reviewed studies reported an increase in birthweight as father's height increased. The magnitude of increase between the lowest group and the highest group was approximately 125-150 g. No studies reported incidences of LBW, PTB, or SGA births. The effect is presumed to be of genetic origin.
Studies of paternal weight and its influence on birthweight revealed contradictory results. Two studies reported an approximate 75- to 100-g increase in birthweight among the highest weight/body mass index group,
All of these studies had very small sample sizes, and a large population-based study is warranted.
Paternal birthweight
Lower paternal birthweight was associated with lower infant birthweight. Paternal LBW was associated with higher risk of an infant with LBW, which is an effect that is similar to maternal LBW. PTB rate was lower for fathers in the LBW birth category. However, this information comes from very few and small studies. Ascertainment issues regarding birthweight of the father should not be ignored in the interpretation of these results. Also, studies from different countries were included in this review; therefore, it may have affected the results because there are different norms for birthweight in different ethnic backgrounds.
Paternal occupational exposure
There are 2 pathways by which paternal exposure contribute to an effect on birth outcomes: either paternal exposure leads to maternal exposure and the effect is exerted or paternal exposure leads to alteration in the germ cell line that leads to either increased infertility or abnormality in conception.
Lead, dioxin, and organic solvents are the more commonly studied exposures. Animal experiments reported lower birthweight after male animal exposure to lead.
Higher and prolonged exposure to lead may be associated with increased risk of LBW births and PTBs; however, ascertainment bias played a major role in included studies. Agent Orange was a widely used herbicide in Vietnam, and contamination of this by dioxin prompted a series of studies that explored the association of paternal exposure with birth outcomes.
Paternal occupational exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin and birth outcomes of offspring: birth weight, preterm delivery, and birth defects.
Paternal occupational exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin and birth outcomes of offspring: birth weight, preterm delivery, and birth defects.
Paternal occupational exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin and birth outcomes of offspring: birth weight, preterm delivery, and birth defects.
The biologic rationale for such an effect remains to be understood. Less than a high school education was associated with a higher rate of PTB, and education less than college level was associated with higher odds of LBW births. Paternal low education level may be a marker for other underlying health determinants such as poverty, adverse home environment, and associated stress.
There was significant heterogeneity in the reported studies that assessed paternal factors and its association to offspring outcomes. Keeping this in mind, we planned for a systematic review from the onset and not metaanalyses. Major clinical heterogeneities among studies were identified for inclusion of the studies (single institution-based study vs national sample) and for adjustment of confounders (different studies controlled for different confounders). A number of studies identified associations in univariate analyses; however, when confounders such as maternal factors were considered, the association became nonsignificant. The major issues with parental occupational exposure were ascertainment of a bias for exposure and recall bias.
Strengths of this systematic review lie in the focused question, comprehensive and extensive literature search, and inclusion of studies from various countries. However, there are limitations of this review. First, there was heterogeneity in studies even within various exposure categories. Second, data on paternal factors are not collected regularly because fathers may not be present at prenatal visits and data are collected from mothers. The information that is collected could be subject to inaccuracies. Additionally, occupational exposure data were obtained from administrative databases or calculated in terms of duration or intensity of exposure. This is subject to ascertainment bias. Additionally, no detailed information regarding the timing of exposure and the protection used by fathers were reported. Third, various studies adjusted for different confounders in their analyses, and some studies did not adjust for confounders. It is important to adjust for various maternal confounders to assess the independent effect of paternal exposure. This limited our ability to conclude anything definitively about any of the factors. Fourth, publication bias should be considered. There is a possibility that, if the studies reported no association for any of the paternal exposures, there may have been less likelihood of publication. Fifth, we restricted our search to English language publications primarily because of the scope of information that may or may not be available in other languages. Literature searchers have found minimal difference with the inclusion of non-English language articles.
We have not included gray literature, abstracts, and proceedings because the quality of such studies cannot be assessed adequately. Finally, none of the studies used any objective measure to confirm paternity; the data on paternity were obtained from maternal history in all studies.
Implications
There are clinically significant and important implications of these results. Pregnancies that were conceived from younger and advanced age fathers, short-statured fathers, fathers who were LBW at birth, and fathers with less than a college education could be considered to be at a greater risk. This information can be used during counseling and education sessions.
Conclusion
In this comprehensive systematic review of paternal factors, associations were identified for paternal age, height, and LBW. Higher paternal birthweight resulted in heavier offspring. Heavy and prolonged lead exposure may be associated with LBW. Paternal weight, body mass, and workers with occupational exposures to herbicides, plant work, and wood were not associated with birth outcomes. Low paternal education may be associated with LBW/PTB births; however, further studies are warranted.
Acknowledgments
We sincerely thank Elizabeth Uleryk, Chief Librarian at the Hospital for Sick Children, Toronto, for her contribution in developing search strategy and running searches on a periodic basis, for which she did not receive any compensation. Contributors: guarantor: Shah PS; grant concept and design: all members of the group; study concept and design: Shah PS; acquisition of data, analysis, and interpretation of data, drafting of the manuscript: Shah PS; critical revision of the manuscript for intellectual content: Shah PS and all members of group. Members of the Knowledge Synthesis Group on determinants of LBW/preterm births: Prakesh S Shah, University of Toronto, Toronto, Ontario, Canada; Arne Ohlsson, University of Toronto, Ontario, Canada; Sarah D McDonald, McMaster University, Hamilton, Ontario, Canada; Eileen Hutton, McMaster University, Hamilton, Ontario, Canada; Vibhuti Shah, University of Toronto, Toronto, Ontario, Canada; Joseph Beyene, University of Toronto, Toronto, Ontario, Canada; Corine Frick, University of Calgary, Calgary, Alberta, Canada; Fran Scott, University of Toronto, Toronto, Ontario, Canada; Kellie E Murphy, University of Toronto, Ontario, Canada; Christine Newburn-Cook, University of Alberta, Edmonton, Alberta, Canada; Victoria Allen, Dalhousie University, Halifax, Nova Scotia, Canada.
Appendix
Tabled
1Quality assessment tool
Bias
None
Low
Moderate
High
Cannot tell
Selection
Consecutive unselected population
Sample selected from large population; selection criteria not defined
Sample selection ambiguous; sample may be representative
Sample selection ambiguous; sample likely not representative
NA
Sample selected from general population rather than a select group
A select group of population (eg, based on race, ethnicity, residence)
Eligibility criteria not explained
A very select population was studied, which made it difficult to generalize findings
NA
Rationale for case and control selection explained
Rationale for case and control subjects not explained
NA
Follow-up or assessment time explained
Follow-up or assessment time not explained
NA
Exposure assessment
Direct questioning (interview) or completion of survey by mother at the time of exposure or close to the time of exposure
Assessment of exposure from global dataset
Extrapolating data from population exposure sample (with some assumptions) and not direct assessment at any time
Indirect method of assessment (obtaining data from others and not from mother or father)
Paternal occupational exposure to 2,3,7,8-tetrachlorodibenzo-p-dioxin and birth outcomes of offspring: birth weight, preterm delivery, and birth defects.
Effect of paternal alcohol consumption before conception on infant birth weight: ALSPAC Study Team: Avon Longitudinal Study of Pregnancy and Childhood.
This study was supported by funding from the Canadian Institute of Health Research (CIHR) Knowledge Synthesis/Translation Grant no. KRS 86242 . The CIHR has played no role in the analyses, the writing of the report, the interpretation of data, or the decision to submit the manuscript.
Studies have rarely considered the impact of paternal factors on perinatal outcomes generally or on racial differences therein. Shah et al1 have produced a literature review that begins to delve into the contribution of fathers to the risk of adverse birth outcomes. Paternal exposures that were selected for inclusion in their database search were father's age, anthropometry (eg, height, weight), self birthweight, occupational exposures, and education. After a systematic search of the literature, the authors identified the father's age, current weight, and his own birthweight as risk factors for low birthweight and suggest that paternal occupation and education may be important but have not consistently been reported to increase the risk of adverse birth outcomes.
00952-1&id=gr1.jpg){kind=link}
00952-1&id=gr2.jpg){kind=link}