Risk of infertility following pelvic angiographic embolization in female patients with pelvic fractures: A nationwide population-based cohort study in Taiwan.

Among patients with multiple blunt traumatic injuries, 5%–16% sustain trauma to the pelvic ring, resulting in a mortality rate of 11%–54%, primarily due to haemorrhagic shock [ 1 – 3 ]. Rapid haemostasis is necessary to avoid the “lethal triad” of hypothermia, coagulopathy, and acidosis secondary to hypotension and tissue hypoperfusion [ 4 ]. Embolization is commonly used to control arterial bleeding after pelvic trauma. Selective angiogram with injection of each hypogastric artery is recommended. However, the potential for distal bleeding from collateral circulation can be a significant problem. In patients who are haemodynamically unstable with multiple areas of contrast extravasation, nonselective proximal embolization of the anterior, posterior, or occasionally the hypogastric artery with Gelfoam (Pfizer, New York, USA) can be a lifesaving procedure. Serious complications have been reported after pelvic angiographic embolization (PAE) including uterine and bladder necrosis, paresis, buttock ischaemia, and impotence [ 5 – 7 ]. In addition, many researchers have evaluated the relationship between PAE for postpartum haemorrhage and fertility. They concluded that embolization is a safe and effective non-surgical method that does not alter subsequent fertility [ 8 , 9 ]. However, based on our observation and clinical experience, radiation exposure in these patients is high due to repeated follow-up radiographic exams after PAE. The association between PAE in pelvic fractures and infertility is currently under investigation. The definition of infertility is failure of a couple to conceive after one year of normal regular intercourse without contraception in women less than 35 years of age or after six months of normal regular intercourse without contraception in women 35 years and older [ 10 ]. Unexplained infertility accounts for 15% of female infertility [ 11 , 12 ]. In this study, we hypothesized that, due to radiation exposure, infertility may be increased in female patients with pelvic fracture treated with PAE. Therefore, we performed a large-scale nationwide retrospective cohort study using the Taiwan National Health Insurance Research Database (NHIRD) to evaluate the risk of infertility following PAE in female pelvic fracture patients.

A total of 36 and 18,029 patients were included in the case and control cohorts, respectively. The mean age of the case and control cohort was 38.59±15.10 and 37.10±13.45 years, respectively. In the case cohort, the maximum age was 59.70 years and minimum age was 17.60 years. In the control cohort, the maximum age was 60.00 years and minimum age was 15.00 years. Fracture locations and proportion of patients in the two age groups (15–35 years and 36–60 years) were not significantly different between the two cohorts. Proportion of patients with inflammatory disease of the ovary, fallopian tube, pelvic cellular tissue, peritoneum, or with endometriosis was higher in the case cohort than in the control cohort ( p < 0.001). The proportion of patients with inflammatory disease of the uterus, cervix, vagina and vulva, hypertension, hypertriglyceridemia, hypercholesterolemia, and coronary artery disease was not statistically different between the two cohorts. Eight of 36 patients in the case cohort had a diagnosis of diabetes mellitus versus 561 of 18,029 patients in the control cohort ( p < 0.001). There was no significant difference in the ISS between the two cohorts ( p = 0.341). The proportion of patients receiving blood transfusion was higher in the case cohort than in the control cohort ( p < 0.001; Table 1 ). p -value (category variable: Chi-square/Fisher exact test). ISS = Injury Severity Score The average follow-up in the case cohort was 3.90±3.56 years; maximum follow-up period was 13.10 years; and minimum follow-up period was 0.01 years. In the control group, average follow-up time was 2.17±3.30 years; maximum follow-up period was 13.94 years; and minimum follow-up period was 0.01 years. The incidence of infertility in the case cohort was 782.70 per 10,000 person–years. In the control cohort, the incidence was 4.85 per 10,000 person–years ( Fig 1 , Table 2 ). PYs = person–years; Adjusted HR = adjusted hazard ratio: adjusted for all variables listed in the table; CI = confidence interval; ISS = Injury Severity Score After adjusting for possible confounding factors, the incidence of infertility in the case cohort was nearly 30.7-fold higher than that in the control cohort (HR = 30.676, 95% CI = 10.643–70.109). Table 2 shows the estimated HRs of demographic–specific and comorbidity-specific variables for both study cohorts. Patients in both age groups (15–35 years and 36–60 years) had a higher incidence of infertility in the case cohort than in the control cohort (adjusted HR = 49.906 and 23.036, 95% CI = 15.177–64.099 and 11.023–59.108, respectively). Patients without inflammatory disease of the ovary, fallopian tube, pelvic cellular tissue, peritoneum, and endometriosis showed higher risks of developing infertility than patients with these diseases, with adjusted HRs of 108.104 (95% CI = 22.625–516.523) for inflammatory disease of the ovary, fallopian tube, pelvic cellular tissue, and peritoneum and 33.127 (95% CI = 10.496–49.802) for endometriosis. We conducted sensitivity analyses to evaluate the associations between angiographic embolization and the risk of developing infertility according to the duration of follow-up ( Table 3 ). These findings implied that, compared with the control cohort, the case cohort was associated with a significantly higher risk of developing infertility as the follow-up duration was increased. Particularly, patients treated with PAE had a significantly greater incidence of developing infertility when the follow-up period was longer than 2 years; adjusted HRs were 54.496 (95% CI = 15.442–91.176) for follow-up periods of 2–3 years and 38.242 (95% CI = 10.396–80.771) for follow-up periods of >3 years. PYs = person–years; Ratio = case rate ÷ control rate; Adjusted HR = adjusted hazard ratio: adjusted for all variables listed in Cox regression table; CI = confidence interval

Female pelvic fracture patients treated with PAE might be associated with a higher risk of developing infertility in Taiwan. Physicians should be aware of the link and inform patients of this risk prior to arterioembolization.

Taiwan began its National Health Insurance (NHI) program, a single-payer and universal insurance plan, with 97% coverage rate among clinics and hospitals in 1996. Almost 99% of the population of Taiwan received health care provided by the NHI [ 13 ] in 1998. The NHI created the NHIRD for researchers in Taiwan, which has been widely applied in epidemiologic and clinical studies [ 14 – 16 ]. The NHIRD includes the annual registration files and original claims data for reimbursement, and is managed by the National Health Research Institutes (NHRI). We used data from the NHIRD, provided by the Bureau of National Health Insurance of the Department of Health in Taiwan, from the period of 1997–2010. In this study, we collected disease histories from inpatient registry files of the NHIRD. Personal identification information is encrypted and unavailable before the data are released for research in order to protect patient privacy. The disease diagnoses were acquired from the International Classification of Diseases, Ninth Revision, Clinical Modification (ICD–9–CM). Our study was approved by the Institutional Review Board II of the Tri-Service General Hospital, National Defense Medical Center (approval number 1-105-05-050). The study protocol was conducted following the Helsinki Declaration of 1975, as revised in 1983. We performed a population-based retrospective cohort study to clarify the relationship between pelvic fracture treated with angiographic embolization in female patients and the risk of infertility. Inclusion criterion was the diagnosis of pelvic fracture (ICD–9–CM808) from 1997–2010. Exclusion criteria were male gender, age 60 years, history of prior surgery of the genital organs (ICD–9–CM OP65.3–OP65.7, OP65.9, OP66-69, except OP68.1, OP69.93), and history of radiotherapy (ICD–9–CM V580). Five gynaecologic and obstetric specialists determined the inclusion and exclusion criteria. The index date was set as the date of PAE. A total of 18,065 patients were included. Pelvic fracture patients treated with angiographic embolization (n = 36) were identified and classified as the case (ICD–9–CM 444.8) cohort. The control cohort consisted of pelvic fracture patients without angiographic embolization (n = 18,029). Follow-up period was terminated upon the development of infertility (ICD–9–CM 628), withdrawal from the insurance program, or by December 31, 2010 ( Fig 1 ). Anatomic locations of pelvic fractures were recorded for each patient. ICD–9–CM, International Classification of Diseases, Ninth Revision, Clinical Modification. Because a variety of factors can influence infertility, such as age, infection of the genital organs, and severity of concomitant disease [ 17 , 18 ], we considered the following confounding factors: age group (15–35 years, 36–60 years); inflammatory disease of the ovary, fallopian tube, pelvic cellular tissue, and peritoneum (ICD–9–CM 614); inflammatory disease of the uterus (ICD–9–CM 615); inflammatory disease of the cervix, vagina, and vulva (ICD–9–CM 616); endometriosis (ICD–9–CM 617); hypertension (ICD–9–CM 401–5); diabetes mellitus (ICD–9–CM 250); hypertriglyceridemia (ICD–9–CM 275.41); hypercholesterolemia (ICD–9–CM 272.4); coronary artery disease (ICD–9–CM 410–414); history of blood transfusion; and the Injury Severity Score (ISS, ICD–9–CM959.99). To demonstrate the differences between the case and control cohorts, the count and percentage for category variables (locations of pelvic fracture; age group; inflammatory disease of the ovary, fallopian tube, pelvic cellular tissue, and peritoneum; inflammatory diseases of the uterus; inflammatory disease of the cervix, vagina, and vulva; endometriosis; hypertension; diabetes mellitus; hypertriglyceridemia; hypercholesterolemia; coronary artery disease; blood transfusion; ISS) are presented. The chi-square test or Fisher exact test for categorical variables was used to statistically examine the differences between the two cohorts. The cumulative risk of infertility incidence and demographic-specific and comorbidity-specific infertility incidence for the case and control cohorts were compared using a Cox’s proportional hazards regression model adjusted for potential confounding factors to estimate the hazard ratios (HRs) and 95% confidence intervals (CIs) for the case cohort. We used SAS 9.3 software (SAS Institute, Cary, NC, USA) to manage and analyse the data. Significance level was set at <0.05, and all tests were two-sided.

(DOCX) Click here for additional data file.