International Journal of Surgery 34 (2016) 88e95
Contents lists available at ScienceDirect
International Journal of Surgery
journal homepage: www.journal-surgery.net
Review
Risk of colorectal cancer with hysterectomy and oophorectomy:
A systematic review and meta-analysis
Ganfeng Luo 1, Yanting Zhang 1, Li Wang, Yuanwei Huang, Qiuyan Yu, Pi Guo, Ke Li*
Department of Public Health, Shantou University Medical College, No.22 Xinling Road, Shantou, Guangdong, 515041, China
h i g h l i g h t s
Risk of CRC was increased for women undergoing hysterectomy or oophorectomy.
Given that 300,000 women without susceptibility genes for ovarian cancer or metrocarcinoma undergo oophorectomy or hysterectomy every year, the
association of oophorectomy or hysterectomy with increased morbidity of CRC in the entire population has implications for public health guidance.
Lacking randomized controlled trials, these high-quality cohort studies with large size and high follow-up rate of long-term follow-up offer a good
method to assess these associations.
This meta-analysis is the first to evaluate these controversial results.
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 26 August 2016
Background: Colorectal cancer (CRC) is the second most commonly diagnosed cancer worldwide in
females. Sex hormones may play a protective effect in CRC pathogenesis. Ovarian sex steroid levels are
reduced in premenopausal women after hysterectomy. Prospective studies have revealed an 80%
decrease in serum oestradiol levels after bilateral oophorectomy in premenopausal women. We aimed
to elucidate the relationship between hysterectomy or oophorectomy and risk of CRC.
Methods: We estimated relative risk (RR) and 95% confidence intervals (95% CIs) with the metaanalysis. Cochran's Q test and Higgins I2 statistic were used to check for heterogeneity. Subgroup
and sensitivity analyses were performed as were Egger's and Begg's tests and the “trim-and-fill”
method for publication bias analysis.
Results: Risk of CRC was increased 30% for women undergoing oophorectomy relative to the general
population and 24% with hysterectomy relative to no surgery. The risk was increased 22% with hysterectomy with bilateral salpingoo-ophorectomy as compared with simple hysterectomy. On subgroup
analysis, risk of rectal cancer was increased 28% and colon cancer 19% with hysterectomy. Europeans
seem to be sensitive to the risk of CRC, with 27% increased risk after hysterectomy. The risk of CRC
after oophorectomy gradually increased with age at oophorectomy. The risk was greater with bilateral
oophorectomy, with 36% increased risk, than unilateral oophorectomy, with 20% increased risk. Risk
was increased 66% with time since oophorectomy 1e4 years as compared with 5e9 and 10 years.
Conclusions: Risk of CRC was increased for women undergoing hysterectomy or oophorectomy. Women
with susceptibility genes for ovarian cancer or metrocarcinoma should choose oophorectomy or hysterectomy. For women not at high risk for these cancers, oophorectomy or hysterectomy should not be
recommended for increasing the subsequent risk of CRC.
© 2016 IJS Publishing Group Ltd. Published by Elsevier Ltd. All rights reserved.
Keywords:
Hysterectomy
Oophorectomy
Colorectal cancer
Relative risk
Meta-analysis
1. Background
* Corresponding author.
E-mail addresses: luoganfeng1991@126.com (G. Luo), zhangyanting1992@126.
com (Y. Zhang), wangli3740@126.com (L. Wang), hywwell@126.com (Y. Huang),
qy_yu1990@126.com (Q. Yu), guopi.01@163.com (P. Guo), keli1122@126.com (K. Li).
1
Ganfeng Luo and Yanting Zhang contributed equally to this study and share first
authorship.
Colorectal cancer (CRC), the second most commonly diagnosed
cancer and third leading cause of cancer deaths worldwide in females, accounts for an important proportion of the global burden of
http://dx.doi.org/10.1016/j.ijsu.2016.08.518
1743-9191/© 2016 IJS Publishing Group Ltd. Published by Elsevier Ltd. All rights reserved.
G. Luo et al. / International Journal of Surgery 34 (2016) 88e95
cancer incidence and mortality rates [1]. Primary prevention of CRC
should be a preferential task of public health. Smoking, physical
inactivity, overweight and obesity, red and processed meat consumption, and excessive alcohol consumption play a part in CRC
pathogenesis, as do sex hormones, especially estrogen, and estrogen therapy is used for protection [2] [3]. Morbidity and mortality
are higher in men than women [4].
Estrogen, especially oestradiol, has revealed this phenomenon
by several mechanisms that include reduced secondary bile acid
production, reduced circulating insulin like growth factor-I, stimulating humoral and cell-mediated immune response and inhibiting cell proliferation of colorectal tumors by binding to the estrogen
receptor-like ER-b [2,5e8]. The expression of ER-b is lower in
tumour tissue than normal colonic mucosa and is inversely related
to stage of CRC [9]. Earlier age at natural menopause is related to
increased risk of CRC [10]. As well, two prospective cohort studies
in the general population revealed no association of testosterone
levels and CRC [11,12]. However, androgen deprivation therapy may
increase the risk of CRC [13,14]. Observational and experimental
studies have revealed that exposure to oral contraceptives and
hormone replacement therapy lowers the risk [15e18]. However,
caseecohort and caseecontrol studies have shown conflicting results regarding the risk of CRC and endogenous levels of sex steroids in postmenopausal women [19e23].
Hysterectomy is one of the most frequent gynecologic surgeries
among women. Overall, 90% of hysterectomies are performed
because of benign gynecological conditions such as symptomatic
uterine fibroids, endometriosis or unusual uterine bleeding
[24e26]. In the United States, about 600,000 women undergo
hysterectomy every year [26,24]. In European countries, the
prevalence of hysterectomy is highest in Finland (390/100 000
women of any age) [27] and Denmark (360/100 000 women of any
age) [28]. Hysterectomy weakens ovarian function by damaging
ovarian tissue or compromises the blood supply theoretically, as
was shown in prospective studies of women before and after
simple hysterectomy [29e31]. Premenopausal women after hysterectomy with ovarian preservation have higher hormone contents, lower ovarian sex steroid levels, and earlier menopause than
those without hysterectomy [30,32,33]. The morbidity of breast
cancer is reduced by one third after hysterectomy [34,35]. However, in recent studies, hysterectomy increased the risk of CRC [36],
whereas previous studies found no association of hysterectomy
and CRC risk [37e39].
To reduce the risk of ovarian cancer [40,41] and breast cancer
[42e47], bilateral oophorectomy is recommended for benign lesions. Approximately 300,000 women undergo prophylactic oophorectomy each year in the United States. Prospective studies have
found decreased serum oestradiol levels by 80% after bilateral oophorectomy in premenopausal women [48]. Postmenopausal
woman with ovaries sostenuto secrete abundant testosterone and
androstenedione, which is translated into estrogen peripherally
[49,50]. Androgen content is reduced by 50% after bilateral oophorectomy in both premenopausal and postmenopausal women
[29,48,49,51e53]. However, the relationship between oophorectomy and risk of CRC is still unclear. A positive association was
revealed by a few epidemiologic studies [39,42,54], but others
[55,42] had negative results.
To elucidate the relationship between hysterectomy or oophorectomy and risk of CRC, we performed a systematic review and
meta-analysis to summarize the published epidemiologic evidence.
This meta-analysis is the first to evaluate these controversial
results.
89
2. Materials and methods
2.1. Search strategy and study selection
Two authors independently searched PubMed for articles published in English up to June 16, 2016 by using the following key
words: (“Colonic Neoplasms”[Mesh] OR “Rectal Neoplasms”[Mesh]
OR “Colorectal Neoplasms”[Mesh] OR ((colon[tiab] OR colonic[tiab]
OR rectal[tiab] OR rectum[tiab] OR colorect*[tiab] OR large bowel
[tiab]) AND (cancer*[tiab] OR carcinoma*[tiab] OR adenocarcinoma*[tiab] OR adenoma*[tiab] OR malignan*[tiab] OR tumour*
[tiab] OR tumour*[tiab] OR neoplas*[tiab]))), and (“Ovariectomy”[Mesh] OR ovariectomy[tiab] OR oophorectomy[tiab] OR
“Hysterectomy”[Mesh] OR hysterectomy[tiab] OR “Hysterectomy,
Vaginal”[Mesh]). Titles, abstracts, full texts and reference lists of all
identified reports were reviewed in duplicate by the two authors,
and extracted articles were double-checked. Disagreements were
resolved by discussion among the three authors. Reference lists
from related main studies and review articles were also checked for
additional relevant reports.
2.2. Inclusion and exclusion criteria
Studies were considered eligible if 1) participants were from a
general population (i.e., not a specific disease group); 2) the
exposure of interest was hysterectomy or oophorectomy or their
combination; 3) the control group was defined; 4) the outcome of
interest was the diagnosis of CRC; 5) articles provided adjusted risk
ratio or equivalent risk variables (i.e., hazard ratio, odds ratio), and
corresponding 95% confidence intervals (CIs) or data to calculate
them. We excluded the following: 1) reviews and letters; 2)
duplicate publications; 3) unqualified data; and 4) articles with
participants who had a history of cancer before baseline, a family
history of ovarian cancer or metrocarcinoma, or reproductive surgery after natural menopause. When two articles appeared to
report results with overlapping data, only the data representing the
most recent publication or with the larger sample size were
included in the meta-analysis. No publications were excluded on
the basis of quality, sample size, or other objective criteria relevant
to study design and analysis.
2.3. Quality assessment
We evaluated the quality of all reports included by the
Newcastle-Ottawa quality assessment scale for cohort studies [56]
(http://www.ohri.ca/programs/clinical_epidemiology/nosgen.pdf).
Quality mainly involved 1) selection (representativeness of the
exposed cohort, selection of the non-exposed cohort, ascertainment of exposure, demonstration that outcome of interest was not
present at the study start); 2) comparability (comparability of cohorts on the basis of the design or analysis); and 3) outcome
(assessment of outcome, follow-up long enough for outcomes to
occur, and adequacy of follow-up of cohorts).
2.4. Data extraction
We extracted data on 1) publication details (first author‘s name,
year of publication and study design); 2) baseline characteristics of
the studied population (country, numbers of observation group,
cancer, follow-up time, hormone therapy); (3) surgery detail (surgical method, age at surgery, time since surgery); (4) RR of CRC for
different gynecological surgeries and 95% CIs. If these data were
90
G. Luo et al. / International Journal of Surgery 34 (2016) 88e95
absent, we sent an email to the authors.
2.5. Statistical analysis
We extracted the RRs and their 95% CIs from studies to evaluate
the risk of CRC with different gynecological surgeries, and the log
RR and standard error (log RR) were used to aggregate the RR. A
pooled RR > 1 indicated greater risk for the group with surgery. The
effect of gynecological surgery on risk of CRC was considered statistically significant with 95% CIs for the pooled RR not overlapping
1 in the forest plot. Cochran's Q test and Higgins I2 statistic were
used to check for heterogeneity of combined results [57,58]. Studies
were considered heterogeneous with Pheterogeneity < 0.1 or I2 > 50%.
With heterogeneity, the random-effects model (DerSimonian-Laird
method) was applied [59]; otherwise, the fixed-effects model
(Mantel-Haenszel method) was used [60]. We also investigated
potential sources of heterogeneity by subgroup analyses of study
location, age at surgery, time since surgery to enrollment, and type
of surgery. Sensitivity analyses involved excluding studies one by
one. We investigated publication bias by the Egger's linear regression test [61] and Begg's methods [62] as well as by evaluating
funnel plots. The effect of potential publication bias on risk
assessment was further evaluated by the Duval and Tweedie
nonparametric “trim-and-fill” method [63]. Two-sided p < 0.05
was considered statistically significant. All analyses involved use of
STATA 12.0 (STATA Corp., College Station, TX, USA).
was associated more with bilateral oophorectomy (RR ¼ 1.36, 95%
CI 1.29e1.42; Pheterogeneity ¼ 0.336) than unilateral oophorectomy
(RR ¼ 1.20, 95% CI 1.13e1.28; Pheterogeneity ¼ 0.328). Subgroup analyses revealed risk of CRC associated with time since surgery 1e4
years (RR ¼ 1.66, 95% CI 1.55e1.77; Pheterogeneity ¼ 0.251), 5e9 years
(RR ¼ 1.16, 95% CI 1.08e1.24; Pheterogeneity ¼ 0.468), and 10 years
(RR ¼ 1.28, 95% CI 1.24e1.32; Pheterogeneity ¼ 0.872).
3.4. CRC risk with hysterectomy with bilateral salpingooophorectomy versus simple hysterectomy
Fig. 4 shows the results for 5 studies, indicating no heterogeneity (I2 ¼ 0%, P ¼ 0.744). The pooled RR was 1.22 (95% CI 1.06e1.40,
P < 0.001) by the fixed-effects model, which suggested that hysterectomy with bilateral salpingoo-ophorectomy was associated
with risk of CRC, with 22% greater risk than with simple hysterectomy. Subgroup analyses revealed hysterectomy with bilateral
salpingoo-ophorectomy associated with risk of CRC with age <45
years (RR ¼ 1.31, 95% CI 1.01e1.69; Pheterogeneity ¼ 0.708).
3.5. Sensitivity analysis
3. Results
On sensitivity analysis, eliminating any one study in each group
did not predominantly change the overall RR (Figs. S1, S2, S3),
which indicates the reliability of our results for hysterectomy
relative to no surgery, oophorectomy relative to the general population and hysterectomy with bilateral salpingoo-ophorectomy
relative to simple hysterectomy.
3.1. Study characteristics
3.6. Publication bias
We found 1032 articles for the association of gynecological
surgery and risk of CRC. After manually screening the full text of
articles, 1025 articles were eliminated (Fig. 1). The final metaanalysis involved 7 articles (19 studies) [36e39,42,55,64]. The primary characteristics of eligible studies are in Table S1. The 19
studies included 30,7958 participants from Sweden, America and
Finland. All were cohort studies and had high-quality designs
(Table S2). Meta-analysis results are presented in Table 1.
Publication bias of the included studies was detected by Begg's
and Egger's linear test and summarized in Table 2. Funnel plots
were asymmetric (Fig. S4), with P ¼ 0.036 from Egger's test for
hysterectomy relative to no surgery, which indicated publication
bias. However, the pooled point estimation and 95% CI did not
change after the trim-and-fill method (Table S3), so the metaanalysis results for hysterectomy remained stable despite publication bias.
For oophorectomy relative to the general population and hysterectomy with bilateral salpingoo-ophorectomy relative to simple
hysterectomy, Begg's and Egger's test revealed no publication bias
(P-Begg's ¼ 0.174 and 1.000, P-Egger's ¼ 0.738 and 0.664, respectively). The shape of the funnel plot was not asymmetric (Figs. S5,
S6), for no evidence of publication bias.
3.2. CRC risk with hysterectomy versus no surgery
Fig. 2 shows the results for 10 studies, suggesting heterogeneity
(I2 ¼ 53.1%, P ¼ 0.024), which rationalized further exploration to
disclose factors leading to this heterogeneity. The pooled RR was
1.24 (95% CI 1.17e1.32, P < 0.001) by the random-effects model,
which suggested that hysterectomy was associated with risk of CRC,
with 24% increased risk as compared with no surgery. Subgroup
analyses revealed that hysterectomy was associated with risk of
colon cancer (RR ¼ 1.19, 95% CI 1.03e1.36; Pheterogeneity ¼ 0.007) and
rectal cancer (RR ¼ 1.28, 95% CI 1.19e1.37; Pheterogeneity ¼ 0.934),
and for Europeans (RR ¼ 1.27, 95% CI 1.20e1.34;
Pheterogeneity ¼ 0.066).
3.3. CRC risk with oophorectomy relative to the general population
Fig. 3 shows the results for 4 studies, indicating no heterogeneity (I2 ¼ 0%, P ¼ 0.837). The pooled RR was 1.30 (95% CI 1.27e1.34,
P < 0.001) by the fixed-effects model, which suggested that oophorectomy was associated with risk of CRC and the risk was 30%
higher than for the general population. Subgroup analyses showed
that oophorectomy was associated with risk of CRC with age 60e85
years (RR ¼ 1.49, 95% CI 1.41e1.57; Pheterogeneity ¼ 0.461), 50e59
years (RR ¼ 1.32, 95% CI 1.27e1.37; Pheterogeneity ¼ 0.802) and 40e49
years (RR ¼ 1.26, 95% CI 1.21e1.37; Pheterogeneity ¼ 0.995). Risk of CRC
4. Discussion
CRC is the second most commonly diagnosed cancer worldwide
in females. Sex hormones may have a protective effect in CRC
pathogenesis. We aimed to elucidate the relationship between
hysterectomy or oophorectomy and risk of CRC in this systematic
review and meta-analysis. Risk of CRC was increased for women
undergoing oophorectomy relative to the general population and
with hysterectomy relative to no surgery. The risk was increased
with hysterectomy with bilateral salpingoo-ophorectomy as
compared with simple hysterectomy and with oophorectomy by
age at oophorectomy. Bilateral oophorectomy was a stronger predictor of risk than unilateral oophorectomy. CRC risk was increased
with time 4 years after oophorectomy than later years. Our
findings might help with decision making related to these surgeries
in terms of CRC risk.
Oophorectomy or hysterectomy may be biologically associated
with increased risk of CRC. According to rat and mouse experiments, oophorectomy accelerated colorectal carcinogenesis, and
G. Luo et al. / International Journal of Surgery 34 (2016) 88e95
91
Fig. 1. Flow chart of selecting studies for meta-analysis.
oestrogen exposure was protective [65e68].
Among postmenopausal women, androgen levels are greatly
decreased in women who undergo simple hysterectomy as
compared with natural menopause [29]. In rat and mouse models,
androgen treatment has a protective effect [69,70]. Androgen
deprivation therapy may elevate the risk of CRC [13,14].
In this meta-analysis, we found that risk of CRC was increased
30% for women undergoing oophorectomy and was increased 24%
with hysterectomy. Hysterectomy with bilateral salpingoo-ophorectomy was associated with 22% increased risk as compared with
simple hysterectomy.
Our subgroup analysis findings are interesting, although with
the small sample sizes, the consequences are less robust than the
main analysis, particularly age at surgery. Hysterectomy was associated with 28% increased risk of rectal cancer, higher than the 19%
increased risk of colon cancer. Europeans seem more sensitive to
risk of CRC, with 27% increased risk after hysterectomy as compared
with Americans. We found no significant association between
hysterectomy and risk of CRC by age at hysterectomy. An explanation might be insufficient studies for statistical power. Compared
with the general population, women with oophorectomy had
increased risk of CRC with increased age at oophorectomy. This
finding might be due to the effect of exogenous oestrogen treatment on reducing the risk of CRC and younger groups using oestrogen substitute after oophorectomy more frequently than older
women. Bilateral oophorectomy with 36% enhanced risk was a
stronger predictor than unilateral oophorectomy, with 20% risk,
which strengthened the primary conclusion of oophorectomy. Risk
of CRC was greater by 66% at 1e4 years after oophorectomy than at
5e9 and 10 years after. An explanation might be the rapid
decrease in hormone level after oophorectomy, which was the most
sensitive to risk of CRC, and with hormone therapy, the hormone
level increased again. Compared to simple hysterectomy, with
bilateral salpingoo-ophorectomy, age at hysterectomy did not alter
the risk because of small study size.
Our study has a number of strengths. First, sensitivity and
publication bias analyses revealed that our results were stable.
Second, positive results reported in abundant epidemiology studies
have been ignored for not evaluating the effect of other cogynecological procedures and for not estimating time since gynecological operation and age at surgery to solve the problem of
detection bias. However, our meta-analysis indicates that the
conclusions are not affected by these factors. Third, lacking randomized controlled trials, these high-quality cohort studies offer a
92
G. Luo et al. / International Journal of Surgery 34 (2016) 88e95
Table 1
Summary of meta-analysis results.
Surgical method
No. of studies
Model
RR (95%CI)
I2 (%)
Pheterogeneity
10
Random
1.24 (1.17,1.32)
53.1
0.024
<45
45e54
55
3
3
3
Fixed
Fixed
Fixed
0.99 (0.78,1.25)
1.07 (0.90,1.28)
1.16 (0.87,1.53)
48.6
45.3
0
0.143
0.161
0.866
Colon
Rectum
4
3
Random
Fixed
1.19 (1.03,1.36)
1.28 (1.19,1.37)
75.5
0
0.007
0.934
European
United States
8
2
4
Random
Fixed
Fixed
1.27 (1.20,1.34)
1.04 (0.87,1.25)
1.30 (1.27,1.34)
47.3
0
0
0.066
0.396
0.837
15e39
40e49
50e59
60e85
4
4
4
4
Fixed
Fixed
Fixed
Fixed
1.09 (1.00,1.19)
1.26 (1.21,1.37)
1.32 (1.27,1.37)
1.49 (1.41,1.57)
0
0
0
0
0.903
0.995
0.802
0.461
Unilateral
Bilateral
4
4
Fixed
Fixed
1.20 (1.13,1.28)
1.36 (1.29,1.42)
12.9
11.4
0.328
0.336
1e4
5e9
10
4
4
4
5
Fixed
Fixed
Fixed
Fixed
1.66
1.16
1.28
1.22
26.9
0
0
0
0.251
0.468
0.872
0.744
<45
45e54
55
2
2
2
Fixed
Fixed
Fixed
1.31 (1.01,1.69)
1.01 (0.76,1.35)
1.24 (0.86,1.79)
0
3.5
0
0.708
0.309
0.760
Subgroup
Hysterectomy relative to no surgery
Age at surgery
Cancer type
Area
Oophorectomy relative to general population
Age at surgery
Oophorectomy type
Time since oophorectomy (years)
Hysterectomy with bilateral salpingoo-ophorectomy relative to simple hysterectomy
Age at surgery
(1.55,1.77)
(1.08,1.24)
(1.24,1.32)
(1.06,1.40)
RR, relative risk.
The difference between A and B has statistical significance with its confidence interval without 1 in bold.
Fig. 2. Forest plot of CRC risk with hysterectomy versus no surgery.
good method to assess these associations. For most cohort studies
of our review, the large size and long-term follow-up with high
follow-up rate are prominent advantages. The cancer registration
system has almost complete coverage, and the computerized record linkage procedure is accurate. Thus, technical incompleteness
is unlikely to bias the results of cohort studies. Information on
surgery and important covariates was acquired at baseline and
updated every 2 years. Baseline characteristics; lifestyles,
educational and socioeconomic factors; hormone drug use; and
fertility conditions were adjusted, although residual confounding
could not be eliminated. Women with a family history of gynecology cancer were excluded to reduce confounding by indication for
surgery.
Several limitations of our study must be acknowledged. First,
the lack of significant findings is likely due to the small number of
studies available for analysis, and the subgroup results are less
G. Luo et al. / International Journal of Surgery 34 (2016) 88e95
93
Fig. 3. Forest plot of CRC risk with oophorectomy relative to the general population.
Fig. 4. Forest plot of CRC risk with hysterectomy with bilateral salpingoo-ophorectomy relative to simple hysterectomy.
Table 2
Publication bias of the studies.
Hysterectomy relative to no surgery
Oophorectomy relative to general population
Hysterectomy with bilateral oophorectomy relative to simple hysterectomy
robust than the main analysis, so our conclusions should be carefully considered. Second, for a few cohort studies, although selfreported hysterectomy conditions were not confirmed with medical records, self-reported hysterectomy data and data from hospital
records to verify exposure information were consistent [71,72].
Hysterectomy reporting featured great precision, whereas records
of oophorectomy were less accurate [73]. Incorrect self-reporting of
oophorectomy status can lead to non-differential misclassification
Egger's P value
Begg's P value
0.036
0.738
0.664
0.128
0.174
1.000
error of predictor groups, which may attenuate the associations.
However, self-reporting of oophorectomy was nearly coincident
with the surgical record in the Nurses' Health Study of our review
[72], although other studies have reported lower precision rates
[33,73]. Third, data on endogenous levels of sex hormones after
these surgeries are lacking, so we could not evaluate a doseeresponse relationship. Fourth, when calculating the effect estimates by separating the role of another gynecological surgery from
94
G. Luo et al. / International Journal of Surgery 34 (2016) 88e95
the main surgery, we could not eliminate the potential effect of
other gynecological surgeries and because of insufficient studies
and detailed data. Fifth, most of the women in our studies were
white, so our conclusions may not be applicable to other ethnic
groups.
Women should choose oophorectomy or hysterectomy if they
have known high-penetrance susceptibility genes for ovarian cancer or metrocarcinoma because of the high life-time risk of ovarian
cancer or metrocarcinoma [74]. For women not at high risk of
ovarian cancer or metrocarcinoma, prophylactic gynecological
surgery should not be recommended, because it increases the risk
of CRC. Given that approximately 300,000 US women without these
gene mutations undergo oophorectomy and 600,000 undergo
hysterectomy every year, the association of oophorectomy or hysterectomy with increased morbidity and mortality of CRC in the
entire population has implications for public health guidance.
Despite challenges, additional studies are needed to determine the
biological mechanisms of these associations, and a prospective
randomized controlled trial with prolonged follow-up is essential
to verify our findings.
Ethical approval
All analyses were based on previous published data, thus no
ethical approval and patient consent are required.
Role of the funding source
There was no funding for this study. The corresponding author
had full access to all data in the study and had final responsibility
for the decision to submit for publication.
Author contributions
KL was responsible for the study concept and design. GL, YZ, YH,
QY, LW, and PG acquired data. GL and YZ drafted and wrote the
report. GL performed the statistical analysis.
Conflict of interest
The authors declare no conflict of interest.
Guarantor
Ke Li was the one who accept full responsibility for the work
and/or the conduct of the study, had access to the data, and
controlled the decision to publish.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ijsu.2016.08.518.
References
[1] L.A. Torre, F. Bray, R.L. Siegel, J. Ferlay, J. Lortet-Tieulent, A. Jemal, Global
cancer statistics, 2012, CA a cancer J. Clin. 65 (2) (2015) 87e108, http://
dx.doi.org/10.3322/caac.21262.
[2] A.J. McMichael, J.D. Potter, Reproduction, endogenous and exogenous sex
hormones, and colon cancer: a review and hypothesis, J. Natl. Cancer Inst. 65
(6) (1980) 1201e1207.
[3] W.Y. Cheung, Q. Shi, M. O'Connell, J. Cassidy, C.D. Blanke, D.J. Kerr, et al., The
predictive and prognostic value of sex in early-stage colon cancer: a pooled
analysis of 33,345 patients from the ACCENT database, Clin. colorectal cancer
12 (3) (2013) 179e187, http://dx.doi.org/10.1016/j.clcc.2013.04.004.
[4] J. Ferlay, H.R. Shin, F. Bray, D. Forman, C. Mathers, D.M. Parkin, Estimates of
worldwide burden of cancer in 2008: GLOBOCAN 2008, Int. J. cancer 127 (12)
(2010) 2893e2917, http://dx.doi.org/10.1002/ijc.25516.
[5] P.A. Newcomb, G. Pocobelli, V. Chia, Why hormones protect against large
bowel cancer: old ideas, new evidence, Adv. Exp. Med. Biol. 617 (2008)
259e269, http://dx.doi.org/10.1007/978-0-387-69080-3_24.
[6] H.H. Hsu, W.W. Kuo, D.T. Ju, Y.L. Yeh, C.C. Tu, Y.L. Tsai, et al., Estradiol agonists
inhibit human LoVo colorectal-cancer cell proliferation and migration through
p53, World J. Gastroenterol. 20 (44) (2014) 16665e16673, http://dx.doi.org/
10.3748/wjg.v20.i44.16665.
[7] J.D. Harrison, S. Watson, D.L. Morris, The effect of sex hormones and tamoxifen
on the growth of human gastric and colorectal cancer cell lines, Cancer 63 (11)
(1989) 2148e2151.
[8] E.J. Kovacs, K.A. Messingham, M.S. Gregory, Estrogen regulation of immune
responses after injury, Mol. Cell. Endocrinol. 193 (1e2) (2002) 129e135.
[9] R. Kennelly, D.O. Kavanagh, A.M. Hogan, D.C. Winter, Oestrogen and the colon:
potential mechanisms for cancer prevention, Lancet Oncol. 9 (4) (2008)
385e391, http://dx.doi.org/10.1016/s1470-2045(08)70100-1.
[10] R.K. Peters, M.C. Pike, W.W. Chang, T.M. Mack, Reproductive factors and colon
cancers, Br. J. cancer 61 (5) (1990) 741e748.
[11] D.D. Orsted, B.G. Nordestgaard, S.E. Bojesen, Plasma testosterone in the general population, cancer prognosis and cancer risk: a prospective cohort study,
Ann. Oncol. official J. Eur. Soc. Med. Oncology/ESMO 25 (3) (2014) 712e718,
http://dx.doi.org/10.1093/annonc/mdt590.
[12] Z. Hyde, L. Flicker, K.A. McCaul, O.P. Almeida, G.J. Hankey, S.A. Chubb, et al.,
Associations between testosterone levels and incident prostate, lung, and
colorectal cancer. A population-based study, Cancer Epidemiol. biomarkers
Prev. a Publ. Am. Assoc. Cancer Res. cosponsored by Am. Soc. Prev. Oncol. 21
(8) (2012) 1319e1329, http://dx.doi.org/10.1158/1055-9965.epi-12-0129.
[13] Y. Lu, R. Ljung, A. Martling, M. Lindblad, Risk of colorectal Cancer by subsite in
a swedish prostate Cancer cohort, Cancer control J. Moffitt Cancer Cent. 22 (2)
(2015) 263e270.
[14] S. Gillessen, A. Templeton, G. Marra, Y.F. Kuo, E. Valtorta, V.B. Shahinian, Risk
of colorectal cancer in men on long-term androgen deprivation therapy for
prostate cancer, J. Natl. Cancer Inst. 102 (23) (2010) 1760e1770, http://
dx.doi.org/10.1093/jnci/djq419.
[15] R.T. Chlebowski, J. Wactawski-Wende, C. Ritenbaugh, F.A. Hubbell, J. Ascensao,
R.J. Rodabough, et al., Estrogen plus progestin and colorectal cancer in postmenopausal women, N. Engl. J. Med. 350 (10) (2004) 991e1004, http://
dx.doi.org/10.1056/NEJMoa032071.
[16] J.M. Gierisch, R.R. Coeytaux, R.P. Urrutia, L.J. Havrilesky, P.G. Moorman,
W.J. Lowery, et al., Oral contraceptive use and risk of breast, cervical, colorectal, and endometrial cancers: a systematic review, Cancer Epidemiol. Biomarkers Prev. Publ. Am. Assoc. Cancer Res. cosponsored by Am. Soc. Prev.
Oncol. 22 (11) (2013) 1931e1943, http://dx.doi.org/10.1158/1055-9965.epi13-0298.
[17] J.R. Johnson, J.V. Lacey Jr., D. Lazovich, M.A. Geller, C. Schairer, A. Schatzkin, et
al., Menopausal hormone therapy and risk of colorectal cancer, Cancer Epidemiol. Biomarkers Prev. Publ. Am. Assoc. Cancer Res. Cosponsored Am. Soc.
Prev. Oncol. 18 (1) (2009) 196e203, http://dx.doi.org/10.1158/1055-9965.epi08-0596.
[18] K. Delellis Henderson, L. Duan, J. Sullivan-Halley, H. Ma, C.A. Clarke,
S.L. Neuhausen, et al., Menopausal hormone therapy use and risk of invasive
colon cancer: the California Teachers Study, Am. J. Epidemiol. 171 (4) (2010)
415e425, http://dx.doi.org/10.1093/aje/kwp434.
[19] N. Murphy, H.D. Strickler, F.Z. Stanczyk, X. Xue, S. Wassertheil-Smoller,
T.E. Rohan, et al., A prospective evaluation of endogenous sex hormone levels
and colorectal Cancer risk in postmenopausal women, J. Natl. Cancer Inst. 107
(10) (2015), http://dx.doi.org/10.1093/jnci/djv210.
[20] R.T. Falk, C.M. Dallal, J.V. Lacey Jr., D.C. Bauer, D.S. Buist, J.A. Cauley, et al.,
Estrogen metabolites are not associated with colorectal Cancer risk in postmenopausal women, Cancer Epidemiol. Biomarkers Prev. Publ. Am. Assoc.
Cancer Res. Cosponsored Am. Soc. Prev. Oncol. 24 (9) (2015) 1419e1422,
http://dx.doi.org/10.1158/1055-9965.epi-15-0541.
[21] M.J. Gunter, D.R. Hoover, H. Yu, S. Wassertheil-Smoller, T.E. Rohan,
J.E. Manson, et al., Insulin, insulin-like growth factor-I, endogenous estradiol,
and risk of colorectal cancer in postmenopausal women, Cancer Res. 68 (1)
(2008) 329e337, http://dx.doi.org/10.1158/0008-5472.can-07-2946.
[22] T.V. Clendenen, K.L. Koenig, R.E. Shore, M. Levitz, A.A. Arslan, A. ZeleniuchJacquotte, Postmenopausal levels of endogenous sex hormones and risk of
colorectal cancer, Cancer Epidemiol. Biomarkers Prev. Publ. Am. Assoc. Cancer
Res. Cosponsored Am. Soc. Prev. Oncol. 18 (1) (2009) 275e281, http://
dx.doi.org/10.1158/1055-9965.epi-08-0777.
[23] J.H. Lin, S.M. Zhang, K.M. Rexrode, J.E. Manson, A.T. Chan, K. Wu, et al., Association between sex hormones and colorectal cancer risk in men and
women, Clin. Gastroenterol. hepatology official Clin. Pract. J. Am. Gastroenterological Assoc. 11 (4) (2013) 419e424, http://dx.doi.org/10.1016/
j.cgh.2012.11.012 e1.
[24] J.M. Wu, M.E. Wechter, E.J. Geller, T.V. Nguyen, A.G. Visco, Hysterectomy rates
in the United States, 2003, Obstetrics Gynecol. 110 (5) (2007) 1091e1095,
http://dx.doi.org/10.1097/01.AOG.0000285997.38553.4b.
[25] V.L. Jacoby, A. Autry, G. Jacobson, R. Domush, S. Nakagawa, A. Jacoby,
Nationwide use of laparoscopic hysterectomy compared with abdominal and
vaginal approaches, Obstetrics Gynecol. 114 (5) (2009) 1041e1048, http://
dx.doi.org/10.1097/AOG.0b013e3181b9d222.
[26] M.K. Whiteman, S.D. Hillis, D.J. Jamieson, B. Morrow, M.N. Podgornik,
K.M. Brett, et al., Inpatient hysterectomy surveillance in the United States,
2000-2004, Am. J. obstetrics Gynecol. 198 (1) (2008), http://dx.doi.org/
G. Luo et al. / International Journal of Surgery 34 (2016) 88e95
10.1016/j.ajog.2007.05.039, 34.e1-7.
[27] R. Luoto, J. Kaprio, I. Keskimaki, J.P. Pohjanlahti, E.M. Rutanen, Incidence,
causes and surgical methods for hysterectomy in Finland, 1987e1989, Int. J.
Epidemiol. 23 (2) (1994) 348e358.
[28] T.F. Andersen, M. Madsen, A. Loft, [Regional variations in the use of hysterectomy], Ugeskrift laeger 149 (36) (1987) 2415e2419.
[29] G.A. Laughlin, E. Barrett-Connor, D. Kritz-Silverstein, D. von Muhlen, Hysterectomy, oophorectomy, and endogenous sex hormone levels in older women:
the Rancho Bernardo Study, J. Clin. Endocrinol. metabolism 85 (2) (2000)
645e651, http://dx.doi.org/10.1210/jcem.85.2.6405.
[30] H. Xiangying, H. Lili, S. Yifu, The effect of hysterectomy on ovarian blood
supply and endocrine function, Climacteric J. Int. Menopause Soc. 9 (4) (2006)
283e289, http://dx.doi.org/10.1080/13697130600865774.
[31] C.M. Farquhar, L. Sadler, S.A. Harvey, A.W. Stewart, The association of hysterectomy and menopause: a prospective cohort study, BJOG Int. J. obstetrics
Gynaecol. 112 (7) (2005) 956e962, http://dx.doi.org/10.1111/j.14710528.2005.00696.x.
[32] C.C. Chan, E.H. Ng, P.C. Ho, Ovarian changes after abdominal hysterectomy for
benign conditions, J. Soc. Gynecol. Investigation 12 (1) (2005) 54e57, http://
dx.doi.org/10.1016/j.jsgi.2004.07.004.
[33] W.J. Hehenkamp, N.A. Volkers, F.J. Broekmans, F.H. de Jong, A.P. Themmen,
E. Birnie, et al., Loss of ovarian reserve after uterine artery embolization: a
randomized comparison with hysterectomy, Hum. Reprod. Oxf. Engl. 22 (7)
(2007) 1996e2005, http://dx.doi.org/10.1093/humrep/dem105.
[34] K.L. Irwin, N.C. Lee, H.B. Peterson, G.L. Rubin, P.A. Wingo, M.G. Mandel, Hysterectomy, tubal sterilization, and the risk of breast cancer, Am. J. Epidemiol.
127 (6) (1988) 1192e1201.
[35] T. Hirayama, E.L. Wynder, A study of the epidemiology of cancer of the breast.
II. The influence of hysterectomy, Cancer 5 (1962) 28e38.
[36] J. Segelman, L. Lindstrom, J. Frisell, Y. Lu, Population-based analysis of colorectal cancer risk after oophorectomy, Br. J. Surg. 103 (7) (2016) 908e915,
http://dx.doi.org/10.1002/bjs.10143.
[37] R. Luoto, A. Auvinen, E. Pukkala, M. Hakama, Hysterectomy and subsequent
risk of cancer, Int. J. Epidemiol. 26 (3) (1997) 476e483.
[38] M.M. Gaudet, S.M. Gapstur, J. Sun, L.R. Teras, P.T. Campbell, A.V. Patel, Oophorectomy and hysterectomy and cancer incidence in the Cancer prevention
study-II nutrition cohort, Obstetrics Gynecol. 123 (6) (2014) 1247e1255,
http://dx.doi.org/10.1097/aog.0000000000000270.
[39] D.A. Boggs, J.R. Palmer, L. Rosenberg, Bilateral oophorectomy and risk of
cancer in African American women, Cancer causes control CCC 25 (4) (2014)
507e513, http://dx.doi.org/10.1007/s10552-014-0353-y.
[40] G.A. Colditz, W.C. Willett, M.J. Stampfer, B. Rosner, F.E. Speizer,
C.H. Hennekens, Menopause and the risk of coronary heart disease in women,
N. Engl. J. Med. 316 (18) (1987) 1105e1110, http://dx.doi.org/10.1056/
nejm198704303161801.
[41] W.A. Rocca, B.R. Grossardt, M. de Andrade, G.D. Malkasian, L.J. Melton 3rd,
Survival patterns after oophorectomy in premenopausal women: a
population-based cohort study, Lancet Oncol. 7 (10) (2006) 821e828, http://
dx.doi.org/10.1016/s1470-2045(06)70869-5.
[42] W.H. Parker, M.S. Broder, E. Chang, D. Feskanich, C. Farquhar, Z. Liu, et al.,
Ovarian conservation at the time of hysterectomy and long-term health
outcomes in the nurses' health study, Obstetrics Gynecol. 113 (5) (2009)
1027e1037, http://dx.doi.org/10.1097/AOG.0b013e3181a11c64.
[43] H.B. Nichols, A. Trentham-Dietz, P.A. Newcomb, L.J. Titus, K.M. Egan,
J.M. Hampton, et al., Postoophorectomy estrogen use and breast cancer risk,
Obstetrics Gynecol. 120 (1) (2012) 27e36, http://dx.doi.org/10.1097/
AOG.0b013e31825a717b.
[44] F. Parazzini, C. Braga, C. La Vecchia, E. Negri, S. Acerboni, S. Franceschi, Hysterectomy, oophorectomy in premenopause, and risk of breast cancer, Obstetrics Gynecol. 90 (3) (1997) 453e456.
[45] C. Schairer, I. Persson, M. Falkeborn, T. Naessen, R. Troisi, L.A. Brinton, Breast
cancer risk associated with gynecologic surgery and indications for such
surgery, Int. J. cancer 70 (2) (1997) 150e154.
[46] N. Kreiger, M. Sloan, M. Cotterchio, V. Kirsh, The risk of breast cancer following
reproductive surgery, Eur. J. cancer 35 (1) (1999) 97e101 (Oxford, England :
1990).
[47] D.J. Press, J. Sullivan-Halley, G. Ursin, D. Deapen, J.A. McDonald, B.L. Strom, et
al., Breast cancer risk and ovariectomy, hysterectomy, and tubal sterilization
in the women's contraceptive and reproductive experiences study, Am. J.
Epidemiol. 173 (1) (2011) 38e47, http://dx.doi.org/10.1093/aje/kwq339.
[48] H.L. Judd, Hormonal dynamics associated with the menopause, Clin. obstetrics
Gynecol. 19 (4) (1976) 775e788.
[49] R.H. Fogle, F.Z. Stanczyk, X. Zhang, R.J. Paulson, Ovarian androgen production
in postmenopausal women, J. Clin. Endocrinol. metabolism 92 (8) (2007)
3040e3043, http://dx.doi.org/10.1210/jc.2007-0581.
[50] H.L. Judd, W.E. Lucas, S.S. Yen, Effect of oophorectomy on circulating testosterone and androstenedione levels in patients with endometrial cancer, Am. J.
obstetrics Gynecol. 118 (6) (1974) 793e798.
[51] H.L. Judd, G.E. Judd, W.E. Lucas, S.S. Yen, Endocrine function of the postmenopausal ovary: concentration of androgens and estrogens in ovarian and
[52]
[53]
[54]
[55]
[56]
[57]
[58]
[59]
[60]
[61]
[62]
[63]
[64]
[65]
[66]
[67]
[68]
[69]
[70]
[71]
[72]
[73]
[74]
95
peripheral vein blood, J. Clin. Endocrinol. metabolism 39 (6) (1974)
1020e1024, http://dx.doi.org/10.1210/jcem-39-6-1020.
C.L. Hughes Jr., L.L. Wall, W.T. Creasman, Reproductive hormone levels in
gynecologic oncology patients undergoing surgical castration after spontaneous menopause, Gynecol. Oncol. 40 (1) (1991) 42e45.
S.L. Davison, R. Bell, S. Donath, J.G. Montalto, S.R. Davis, Androgen levels in
adult females: changes with age, menopause, and oophorectomy, J. Clin.
Endocrinol. metabolism 90 (7) (2005) 3847e3853, http://dx.doi.org/10.1210/
jc.2005-0212.
E.J. Jacobs, E. White, N.S. Weiss, Exogenous hormones, reproductive history,
and colon cancer (Seattle, Washington, USA), Cancer causes control CCC 5 (4)
(1994) 359e366.
V.L. Jacoby, D. Grady, J. Wactawski-Wende, J.E. Manson, M.A. Allison,
M. Kuppermann, et al., Oophorectomy vs ovarian conservation with hysterectomy: cardiovascular disease, hip fracture, and cancer in the Women's
Health Initiative Observational Study, Archives Intern. Med. 171 (8) (2011)
760e768, http://dx.doi.org/10.1001/archinternmed.2011.121.
A. Stang, Critical evaluation of the Newcastle-Ottawa scale for the assessment
of the quality of nonrandomized studies in meta-analyses, Eur. J. Epidemiol.
25 (9) (2010) 603e605, http://dx.doi.org/10.1007/s10654-010-9491-z.
J.P. Higgins, S.G. Thompson, J.J. Deeks, D.G. Altman, Measuring inconsistency
in meta-analyses, BMJ 327 (7414) (2003) 557e560, http://dx.doi.org/10.1136/
bmj.327.7414.557 (Clinical research ed).
J.P. Higgins, S.G. Thompson, Quantifying heterogeneity in a meta-analysis,
Statistics Med. 21 (11) (2002) 1539e1558, http://dx.doi.org/10.1002/
sim.1186.
R. DerSimonian, N. Laird, Meta-analysis in clinical trials, Control. Clin. trials 7
(3) (1986) 177e188.
N. Mantel, W. Haenszel, Statistical aspects of the analysis of data from
retrospective studies of disease, J. Natl. Cancer Inst. 22 (4) (1959) 719e748.
M. Egger, G. Davey Smith, M. Schneider, C. Minder, Bias in meta-analysis
detected by a simple, graphical test, BMJ 315 (7109) (1997) 629e634 (Clinical research ed).
C.B. Begg, M. Mazumdar, Operating characteristics of a rank correlation test
for publication bias, Biometrics 50 (4) (1994) 1088e1101.
S. Duval, R. Tweedie, Trim and fill: a simple funnel-plot-based method of
testing and adjusting for publication bias in meta-analysis, Biometrics 56 (2)
(2000) 455e463.
W.H. Parker, D. Feskanich, M.S. Broder, E. Chang, D. Shoupe, C.M. Farquhar, et
al., Long-term mortality associated with oophorectomy compared with
ovarian conservation in the nurses' health study, Obstetrics Gynecol. 121 (4)
(2013) 709e716, http://dx.doi.org/10.1097/AOG.0b013e3182864350.
M.J. Weyant, A.M. Carothers, N.N. Mahmoud, H.L. Bradlow, H. Remotti,
R.T. Bilinski, et al., Reciprocal expression of ERalpha and ERbeta is associated
with estrogen-mediated modulation of intestinal tumorigenesis, Cancer Res.
61 (6) (2001) 2547e2551.
J. Yang, L.J. Xiong, F. Xu, X. Zhao, B. Liu, K.L. Cai, et al., Estrogen inhibits colon
polyp formation by reducing angiogenesis in a carcinogen-induced rat model,
Int. J. Endocrinol. 2013 (2013) 453898, http://dx.doi.org/10.1155/2013/
453898.
J.Y. Guo, X. Li, J.D. Browning Jr., G.E. Rottinghaus, D.B. Lubahn, A. Constantinou,
et al., Dietary soy isoflavones and estrone protect ovariectomized ERalphaKO
and wild-type mice from carcinogen-induced colon cancer, J. Nutr. 134 (1)
(2004) 179e182.
P. Smirnoff, Y. Liel, J. Gnainsky, S. Shany, B. Schwartz, The protective effect of
estrogen against chemically induced murine colon carcinogenesis is associated with decreased CpG island methylation and increased mRNA and protein
expression of the colonic vitamin D receptor, Oncol. Res. 11 (6) (1999)
255e264.
K. Aoki, A. Nakajima, K. Mukasa, E. Osawa, Y. Mori, H. Sekihara, Prevention of
diabetes, hepatic injury, and colon cancer with dehydroepiandrosterone,
J. steroid Biochem. Mol. Biol. 85 (2e5) (2003) 469e472.
J.R. Izbicki, S.R. Hamilton, G. Wambach, E. Harnisch, D.K. Wilker,
G. Dornschneider, et al., Effects of androgen manipulations on chemically
induced colonic tumours and on macroscopically normal colonic mucosa in
male Sprague-Dawley rats, Br. J. cancer 61 (2) (1990) 235e240.
A. Coulter, K. McPherson, S. Elliott, B. Whiting, Accuracy of recall of surgical
histories: a comparison of postal survey data and general practice records,
Community Med. 7 (3) (1985) 186e189.
G.A. Colditz, M.J. Stampfer, W.C. Willett, W.B. Stason, B. Rosner,
C.H. Hennekens, et al., Reproducibility and validity of self-reported menopausal status in a prospective cohort study, Am. J. Epidemiol. 126 (2) (1987)
319e325.
A.I. Phipps, D.S. Buist, Validation of self-reported history of hysterectomy and
oophorectomy among women in an integrated group practice setting,
Menopause 16 (3) (2009) 576e581, http://dx.doi.org/10.1097/gme.0b013e31818ffe28 (New York, NY).
S.A. Gayther, P.D. Pharoah, The inherited genetics of ovarian and endometrial
cancer, Curr. Opin. Genet. Dev. 20 (3) (2010) 231e238, http://dx.doi.org/
10.1016/j.gde.2010.03.001.