From: Tubal ligation, hysterectomy and ovarian cancer: A meta-analysis
Author (Country) | Study Design | Case definition | Covariates | OR, RR, or SIR (95%CI) | Comments |
---|---|---|---|---|---|
NECC 2012 (USA) [personal communication with Dr. Daniel Cramer] | Case-control | Borderline or invasive epithelial ovarian cancer N=2076 | age, study center, BMI , study phase, smoking, family history of ovarian and breast cancers, talc use, OC use, parity, breast feeding, age at menarche, post-menopausal status, use of post-menopausal hormones, hysterectomy | 0.79 (0.66-0.94) | |
Ness et al. 2011 (USA) [11] | Case-control | Invasive or borderline epithelial ovarian cancer | Age, number of pregnancies, race, infertility, family history of | 0.63 (0.51-0.77) | |
N=867 | ovarian cancer, ever use of oral contraceptives, ever use of IUDs, ever use of barriers, vasectomy | ||||
Moorman et al. 2009 (USA) [12] | Case-control North Carolina Ovarian Cancer Study | Invasive epithelial ovarian cancer | Age, parity, age at menarche, duration of OC use, family history of breast/ovarian cancer, BMI | Whites: 0.74 (0.58, 0.94) | |
African-Americans: 0.43 (0.24, 0.80) | |||||
N=746 White cases | |||||
N=111 African-American cases | |||||
Antoniou et al. 2009 (Europe and Canada) [13] | Retrospective Cohort | Ovarian cancer (only BRCA 1/2 carriers) | Age, duration of OC use, parity | BRCA 1/2: 0.43 (0.24, 0.75) | Includes prevalent and incident cases. |
BRCA1: 0.42 (0.22, 0.80) | |||||
N=201 BRCA1 cases | |||||
BRCA2: 0.47 (0.18, 1.21) | Mean difference between age at diagnosis and interview: 6.7 years | ||||
N=52 BRCA2 cases | |||||
Wu et al. 2009 (USA) [37] | Case-control | Invasive and borderline ovarian cancer | Race/ethnicity, age, education, family history of ovarian cancer, menopausal status, use of oral contraceptives, parity | 0.66 (0.47, 0.93) | |
N=609 cases | |||||
Dorjgochoo T. et al. 2009 (China) [14] | Prospective cohort | Ovarian cancer | Age, education, age at menarche, parity, breastfeeding, BMI, physical activity, smoking, menopausal status, family history of cancer, other contraceptive methods. | 1.17 (0.62, 2.26) | Cohort N=66,661 |
N=94 cases | |||||
76.1% participation rate | |||||
Nagle et al. 2008 (Australia) [15] | Case-control | Invasive epithelial endometrioid and clear cell ovarian cancer | Age, education, parity, and hormone contraceptive use | Endometrioid: 0.4 (0.3, 0.7) | 47% participation rate in controls |
Clear cell: 0.7 (0.4, 1.2) | |||||
N=142 endometrioid cases | |||||
N=90 clear cell cases | |||||
Jordan et al. 2008 (Australia) [10] | Case-control | Invasive epithelial serous ovarian cancer | Parity, hormonal contraceptive use, history of breast or ovarian cancer, age, education | Serous (invasive): 0.87 (0.69-1.09) | |
N=627 cases | |||||
Jordan et al. 2007 (Australia) [16] | Case-control | Epithelial benign serous tumors (N=230) and benign mucinous tumors (N=133) | Age, state of residence, education, parity, hormonal contraceptive use, hysterectomy, smoking status | Combined: 1.04 (0.76-1.44) | 65% participation rate in cases, 47% in controls. |
Mucinous: 1.00 (0.61-1.64) | |||||
Serous: 1.08 (0.75-1.57) | |||||
Tworoger et al. 2007 (USA) [17] | Prospective cohort | Incident invasive epithelial ovarian cancer | Age, BMI, parity, smoking history, age at menarche, age at menopause, duration of postmenopausal hormone use, duration of oral contraceptive use | 0.66 (0.50, 0.87) | Update of Hankinson et al. 1993 |
N=612 cases | |||||
McLaughlin JR et al. 2007 (International) [18] | Case-control | Invasive ovarian cancer (only BRCA 1/2 carriers) | Age, mutation type, country of residence, parity, breastfeeding, oral contraceptive use, ethnicity. | BRCA1+2 carriers: 0.78 (0.60, 1.00) | Includes prevalent and incident cases. Results similar when restricted to women interviewed within 3 years of diagnosis. |
BRCA1: 0.80 (0.59, 1.08) | |||||
N=799 cases | BRCA2: 0.63 (0.34, 1.15) | ||||
BRCA1 N=670 BRCA2 N=128 | |||||
BRCA1/2 N=1 | |||||
Modugno et al. 2004 (USA) [9] | Pooled case-control | Epithelial ovarian cancer | Study site, age, family history, duration of oral contraceptive use, parity | 0.63 (0.54, 0.73) | Pooled analysis from four studies. |
N=2098 cases | |||||
Kjaer et al. 2004 (Denmark) [19] | Population-based follow-up study | Invasive ovarian cancer and borderline ovarian tumor | Age and calendar year | Invasive: 0.82 (0.6, 1.0) | Observed number of cancer cases in cohort of women who underwent tubal ligation was compared to the expected number of cases based on the age and calendar year specific rates from the Danish Cancer Registry. |
Borderline: 0.82 (0.5, 1.3) | |||||
N=75 invasive cases | |||||
N=21 borderline cases | |||||
McGuire et al. 2004 (USA) [20] | Case-control | Invasive epithelial ovarian cancer | Age, parity, duration of OC use, race/ethnicity | BRCA 1 carriers: 0.68 (0.25, 1.90) | |
Noncarriers: 0.65 (0.45, 0.95) | |||||
N=36 BRCA1 cases | |||||
N=381 noncarrier cases | |||||
Pike et al. 2004 (Los Angeles, USA) [21] | Case-control | Invasive ovarian cancer | Age, ethnicity, SES, education, family history of ovarian cancer, use of talc, BMI, parity, age at last birth, number of incomplete pregnancies, OC use, menopausal status, age at menopause, hormone replacement therapy | 0.82 (0.53-1.26) | |
N=477 cases | |||||
Rutter et al. 2003 (Israel) [23] | Case-control | Invasive epithelial ovarian cancer or primary peritoneal cancer | Age, ethnicity, parity, years of oral contraceptive use | 0.70 (0.42, 1.18) | Participation rate was 79% for case patients and 66% for controls. |
N=1124 cases | |||||
Wittenberg et al. 1999 (USA) [24] | Case-control | Mucinous and non-mucinous epithelial ovarian cancer | Age at diagnosis, parity, duration of OC use | Mucinous: 0.4 (0.1, 1.9) | 64% participation rate in cases, 72% in controls. Included both borderline and invasive. |
Non-mucinous: 0.6 (0.3, 1.1) | |||||
N=43 mucinous cases | |||||
N=279 non-mucinous cases | |||||
Kreiger et al, 1997 (Canada) [25] | Historical cohort study | Invasive and borderline ovarian cancer | Age, calendar year, length of follow-up | 0.57 p<0.001 | Calculated observed over expected events. |
N=108 observed cases in tubal ligation subcohort | Sensitivity analysis excluding borderline malignancies similar. | ||||
Green, Purdie, et al. 1997 (Australia) [26] | Case-control | Incident, primary epithelial ovarian cancer | Age, education, BMI, parity, OC duration, smoking, family history of ovarian cancer | 0.61 (0.46, 0.85) | 90% participation rate in cases, 73% in controls. |
N=824 cases | |||||
Cornelison et al 1997 (USA) [27] | Case-control | Ovarian cancer N=300 cases | Age , SES, marital status, parity, age at first pregnancy, age at menarche, age at menopause, irregular menses, breast-feeding duration, BMI, OC use | 0.52 (0.31,0.85) | Patient controls with no malignancy or ovarian disease. |
Miracle-McMahill, et al. 1997 (USA) [28] | Prospective Cohort Study | Ovarian cancer mortality | Age, race, BMI, education, family history of ovarian cancer, family history of breast ca, parity, marital status, age at menarche, OC use, ERT, age at menopause, miscarriages smoking status | 0.68 (0.45, 1.03) | |
N=799 ovarian cancer deaths | |||||
Rosenblatt, et al. 1996 (International) [29] | Case-control | Borderline or malignant epithelial ovarian cancer | Age, hospital, year of interview, parity OC use | 0.71 (0.47, 1.08) | No differences observed for borderline and malignant tumors. |
N=385 cases | |||||
Risch et al. 1996 (Canada) [22] | Case-control | Epithelial ovarian cancer | Age, parity, years of OC use, average lactation/pregnancy, total years of ERT, hysterectomy, family history of breast cancer | 0.67 (0.47-0.94) | Invasive and borderline tumors included. |
N=450 cases Borderline | |||||
N=83 Invasive N=376 | |||||
Nandakumar et al. 1995 (India) [30] | Case-control | Ovarian cancer | Age, residential area, parity, age at first birth | 0.25 (0.08, 0.78) | Restricted to ever-married women. Hospital-based controls. |
N=97 cases | |||||
Whittemore et al 1992 (USA) [31] | Pooled case-control | Invasive epithelial ovarian cancer | Age, study, parity, OC use | Hospital-based studies: | Restricted to white women. 6 hospital based studies and 6 population-based studies. |
0.59 (0.38, 0.93) Population-based studies: 0.87 (0.62, 1.20) | |||||
N=2197 cases | |||||
Booth et al 1989 (England) [32] | Case-control | Epithelial ovarian cancer | Age, social class, gravidity, unprotected intercourse | 0.2 (0.1, 0.6) | Cases were less than 65 years old and interviewed within 2 years of diagnosis. Age-matched hospital-based controls. |
N=235 cases | |||||
Shu et al 1989 (China) [33] | Case-control | Invasive epithelial ovarian cancer | Age, education, parity, age at menarche, ovarian cyst | 0.8 (0.4, 1.6) | 89% participation rate in cases, 100% in controls. All <70 years of age. |
N=172 cases | |||||
Koch et al 1988 (Canada) [34] | Case-control | Epithelial ovarian cancer | None | 0.8 (0.5, 1.3) | 47% participation rate in controls. Age-matched, but did not control for age in analyses. |
N=200 cases | |||||
Mori et al 1988 (Japan) [36] | Case-control | Primary epithelial ovarian cancer | Age, parity, marital status, number of induced abortions | 0.5 (0.25, 1.00) | Controls were hospital in-patients with gynecological complaints other than ovarian cancer and OB/GYN outpatients without a malignant ovarian disorder. 100% participation rate in cases and controls. |
N=110 cases | |||||
Koch et al. 1984 (Canada) [35] | Retrospective cohort | Ovarian cancer N=4 cases | Age, nulliparity | 2.4 (0.9, 6.7) | Population who underwent tubal ligation were mental patients. 34% were lost to follow-up. Many underwent the procedure at young ages (i.e. 10-19). Expected rates calculated from a previous retrospective study. Incomplete adjustment for parity. |