Abstract
Objective Recurrent attacks of peritonitis due to familial Mediterranean fever (FMF) may lead to peritoneal adhesions and fallopian tube obstruction. Colchicine, which is the treatment of choice for FMF, may disturb cell division. Secondary amyloidosis, a complication of untreated FMF, may involve the testes and ovaries. Thus, FMF and colchicine may potentially affect fertility and pregnancy in patients with FMF. The aims of the study are to evaluate the causes of infertility and pregnancy outcome in FMF patients and to compare them with 2 groups: non-FMF patients with peritoneal female genital tuberculosis (FGTB) and normal healthy controls.
Methods This is a retrospective study in which FMF patients with reproductive disorders were recruited from the National Center of Medical Genetics and Primary Health Care in Yerevan, Armenia. The patients with FGTB and the healthy controls with reproductive problems were recruited successively from a large gynecology clinic in Yerevan. Genetic analyses for FMF were performed using ViennaLab StripAssay.
Results The FMF group (211 patients) resembles the FGTB group (127 patients) regarding etiologies of infertility. However, in vitro fertilization (IVF) success rate and pregnancy outcome were comparable between the FMF patients and the control group (162 patients). Infertility in patients with FMF was clearly associated with a more severe disease and a lack of adequate colchicine treatment.
Conclusions Colchicine medication and controlled FMF disease do not adversely affect the reproductive system and pregnancy outcome. However, a lack of an appropriate colchicine treatment may cause infertility and poor pregnancy outcome.
Familial Mediterranean fever (FMF) is a hereditary autoinflammatory disease characterized by recurrent attacks of fever and serositis such as peritonitis, pleuritis, synovitis, and pericarditis1. Multiple episodes of peritonitis may lead to peritoneal adhesions, which may cause intestinal obstruction and fallopian tube obstruction2. One of the main complications of untreated FMF is serum amyloid A amyloidosis. In this condition, amyloid fibers are deposited in kidneys, liver, and intestines, and later may involve the cardiovascular system as well3,4. The gene associated with FMF (MEFV) was isolated in 1997 by 2 independent groups5,6. It is located on the short arm of chromosome 16. The treatment of choice for FMF is colchicine, which is able to control FMF attacks and prevents the development of amyloidosis. However, in vitro studies showed that a high dose of colchicine may affect cell division7. Thus, the potential complications of FMF (serosal fibrosis and amyloidosis) and colchicine treatment may affect the reproductive system either by obstruction of the salpinx, causing mechanical infertility, or by defective sperm and oocyte proliferation, leading to difficulties in obtaining pregnancy and normal deliveries8,9,10,11.
In 1970, Mamou investigated the reproductive system in 20 women with FMF and reported that ovarian insufficiency was the cause of infertility in most cases9. Ismajovich, et al found ovulatory disturbances in 13 out of 45 patients with FMF and primary sterility10. Ehrenfeld, et al investigated the fertility and obstetric history of 36 women with FMF11. Thirteen (36%) women had infertility, 6 (46%) of whom had ovulatory dysfunction and 4 (31%) of whom had peritoneal adhesions.
It should be emphasized that most of the above studies described patients who had FMF before the colchicine era, and therefore their fertility and pregnancy outcome were poor. Following the introduction of colchicine for FMF patients, their fertility, pregnancy course, and outcome improved significantly12. In our recent literature review, we looked for large studies dealing with the reproductive system in FMF. We found a few publications, some of which summarized previous findings while others studied a small number of patients, sometimes without controls13,14,15,16. Therefore, we decided to study the reproductive system and pregnancy outcomes in our female patients with FMF in Armenia. We compared them with 2 additional groups: non-FMF patients with peritoneal female genital tuberculosis (FGTB) and healthy controls with reproductive disorders but without FMF, FGTB, or any systemic inflammation or malignancy. The patients with FGTB were chosen due to their peritoneal involvement, which may resemble FMF peritonitis. The healthy individuals served as controls for both groups (FMF and FGTB).
MATERIALS AND METHODS
FMF patients’ group (group 1). This group was chosen from the large dataset of the National Center of Medical Genetics and Primary Health Care in Yerevan, Armenia. During the years 1998–2018, there were 32,000 individuals screened for MEFV mutations. Of this group, we chose successively women in their reproductive years (18–49 yrs) who had a confirmed diagnosis of FMF based upon clinical and genetic criteria, and who were investigated for reproductive problems.
The most frequent complaints about reproductive function were as follows: irregular and painful menstruation, primary and secondary infertility, early and late miscarriage, complications of pregnancy, in vitro fertilization (IVF) failures, problems in ovulation, or recurrent inflammation of the uterus and appendages. Patients who had premature delivery, ovarian apoplexy, or ectopic pregnancy were also included in the study.
Regarding FMF, a severity score was calculated for each patient according to the Tel Hashomer criteria, which includes age of disease onset; attack frequency; presence of arthropathy, erysipeloid rash, or proteinuria; and kidney complications or poor response to colchicine treatment17. There were 3 grades of disease severity: mild (2–5), moderate (6–9), and severe (> 10).
FGTB group (group 2). Since FMF is a prototype of a noninfectious peritoneal inflammation, we thought that patients of the same age and origin who have peritoneal genital tuberculosis with concomitant reproductive problems could form an adequate group for comparison. Two authors (POS and OLS) run a large clinical center for patients with FGTB peritonitis and fertility problems. Age-matched FGTB patients were recruited successively from this clinic during the years 2010–2018.
Healthy controls (group 3). Women with reproductive problems, but without any concomitant systemic disease (especially excluding FMF or FGTB), served as a control group. They were recruited successively from a population of women who visited the same gynecology clinic between the years 2010–2018.
All the patients recruited for the study (3 groups) were interviewed and checked by OLS and POS. For the patients with FMF, a large chart containing demographic, clinical, laboratory, and genetic data was filled (Supplementary Table 1, available from the authors on request). In addition, a full gynecological evaluation was carried out for all the patients analyzed in the study (Supplementary Data 1, available from the authors on request). Informed consent was obtained from all participants. Ethics board approval was obtained from the ethics committee of the Center of Medical Genetics and Primary Health Care (ethics approval number 2/13, 11.02.2018).
Molecular genetic methods for FMF diagnosis. All patients with FMF were followed at the National Center of Medical Genetics and Primary Health Care in Yerevan. Diagnosis of FMF was confirmed using the Tel Hashomer criteria18 and by molecular genetic analyses. We employed the ViennaLab Diagnostics StripAssay, which covered the 12 most common MEFV mutations among Armenians: E148Q, P369S, F479L, M680I (G/C), M680I (G/A), I692del, M694V, M694I, K695R, V726A, A744S, and R761H.
Statistical methods. Statistical analysis was performed using a complex data processing package (SPSS 21.0, IBM Corp.). Mean SD and standard error were used to describe numerical data. For qualitative data, rates and proportions were applied. For comparison of continuous variables, a 2-sided t-test for independent groups was used. For comparison of quantitative outcomes between groups, we used Pearson chi-square test. It was also used to analyze intergroup differences on quantitative features. In case of quantitative limitations, a 2-sided Fisher exact test was used. In all cases, results were considered statistically significant at P ≤ 0.05.
RESULTS
The studied groups. Supplementary Figure 1 (available from the authors on request) depicts the flowchart for recruiting the patients with FMF and reproductive problems. It is shown that out of 32,000 individuals screened in the National Center of Medical Genetics and Primary Health Care in Yerevan between the years 1998–2018, there were 5679 women in their reproduction period that carried MEFV mutations (1 or 2). Of this group, 1102 women were excluded, since they were healthy carriers of a single mutation (heterozygotes) without any clinical manifestation. From the remaining 4577 patients with definite FMF (clinically and genetically), 211 patients were found with reproductive disorders.
During the years 2010–2018, there were 127 patients with FGTB and reproductive disorders recruited for the study. Concomitantly, 162 women with reproductive problems but without FMF, FGTB, or any other systemic disease were recruited as controls.
Demographic feature comparison of FMF, FGTB, and control groups. During the study period, the age of the patients in group 1 ranged between 18 and 45 years (average 21.3 ± 6.4 yrs), while the range was between 20 and 46 years (average of 28.4 ± 7.0 yrs) in group 2. The age in the control group was the highest, with an average of 31.4 ± 7.0 years (P < 0.001). Most of the patients in groups 2 and 3 were over the age of 20 years.
Two hundred and fifty-three out of the 500 patients studied were classified in a low socioeconomic status (SES). Of those, 94 (74%) patients were from the FGTB group and 112 (53%) were from the FMF group, while only 47 (29%) belonged to the control group. Patients with low SES were defined by their lack of high school education, and a lack of either a permanent job or private property. Patients in the middle SES were characterized by living in urban areas and having a higher education, stable income, and access to qualified medical services.
Infertility. According to the World Health Organization (www.who.int) recommendations, infertility is defined as the absence of clinical pregnancy following 12 or more months of regular sexual intercourse without protection. Table 1 shows that 139 (65.9%) out of 211 FMF patients had infertility, of whom 116 (83.5%) had primary infertility. In the FGTB group, infertility was diagnosed in 69 (54.3%) patients, of whom 58 (84.1%) had primary infertility. The control group included 115 (71.0%) infertile patients consisting of 47 (40.9%) patients with primary infertility. The most prevalent cause of infertility was tuboperitoneal. It was diagnosed in 128 infertile patients (92.1%) from the FMF group, in 67 (97.1%) from the FGTB group, and in only 40 (34.8%) in the control group (P = 0.001; Table 1). In 80 (49.4%) patients from the control group, infertility was caused by adhesive processes due to inflammation of the uterus and its appendages, endometriosis, and operative interventions (cystectomy and salpingectomy).
The second cause of infertility in patients from the FMF group was disovulation, found in 77 patients (55.4%). In many cases it was concomitant with tuboperitoneal infertility. Disovulation rate in the FMF group was comparable with that of the control group (59.1%). However, the prevalence of endometrial hyperplasia was significantly lower in the FMF group: 9.4% compared with 19.1% in the control group (P = 0.001).
Permanent infertility was much more common in the FGTB group. It was documented in 12 (17.4%) women, most of whom underwent internal organ resection. Disovulation and uterine hyperplasia were moderately expressed in this group (14.5% and 7.2%, respectively), and were caused by peritoneal tuberculosis. Spontaneous recovery of infertility was observed in higher rate among the normal control (54%) compared with the FGTB (15.9%) and FMF groups (33.1%, P < 0.0001; data not shown).
Thrombophilia was found in 31 patients (14.7%) in group 1, in 5 patients (3.9%) in group 2, and in 73 patients (45.1%) in the control group (Table 2). Thrombophilia was defined as a hypercoagulation state supported by laboratory investigation, including global coagulation tests, identification of thrombophilia markers (thrombin–antithrombin fragments and serum D-dimer), and platelet aggregation. We have also looked for factor-V Leiden deficiency, mutation in the prothrombin gene C20210A, and mutations in the MTHFR gene.
Ovarian and peritoneal amyloidosis and premature ovarian insufficiency were found in the FMF group only. Kidney, liver, and intestinal amyloidosis were detected in 7 (3.3%) patients with FMF. There was no difference in the prevalence of concomitant endocrine diseases among all the groups, including Hashimoto thyroiditis, thyroid nodules, pituitary microadenomas, or hyperprolactinemia.
IVF outcome. The highest rate of successful pregnancy ratio to the absolute number of embryo transfers was seen in the FMF group (26/44, 59.1%) and the lowest was seen in the FGTB group (13/34, 38.2%). This may reflect the relatively lower rate of damaged endometrium (7.1%) in the FMF group compared with 50% in the FGTB group and 23.5% in the control group (data not shown).
For quality assessment of IVF, we used the “take-home baby rate” index, defined as the ratio of the actual number of babies born with survival over 27 days and the number of transfer embryo procedures. The take-home baby index in the FMF group was 36.4%, and it resembled that of the control group (37.5%). The lowest take-home baby rate was in the FGTB group (23.5%). Moreover, the frequency of spontaneous miscarriages in the first trimester was the highest in the FGTB group compared with the FMF and control groups (25% vs 16.7% and 17.6%, respectively; data not shown).
Analysis of pregnancy outcomes and obstetric complications. The main pregnancy outcome analysis included abortions, early termination of pregnancies, and congenital malformations. In the FMF group, 251 pregnancies were documented, of which 149 (59.4%) ended up with a normal outcome (Table 3). Forty pregnancies terminated as early spontaneous miscarriages (15.9%), 4 pregnancies ended as late miscarriages (1.6%), and 4 newborns (1.6%) had congenital malformations. Ectopic pregnancy was observed in 17 patients (6.8%). In the control group, out of 312 pregnancies, 190 (60.9%) ended successfully with live born babies, 51 pregnancies ended with early spontaneous miscarriages (16.3%), and 10 (3.2%) as late miscarriages. Eight (2.6%) newborns had congenital developmental abnormalities.
In the FGTB group, of 133 pregnancies, 79 (59.4%) pregnancies terminated with delivery (Table 3). Fourteen (10.5%) pregnancies ended in early spontaneous miscarriages, 11 (8.3%) ended as late miscarriages. Among late complications of pregnancy, antenatal mortality of fetuses was seen less often than in the control group: 2.3% and 3.5%, respectively. However, fetal hypoxia was significantly higher in the FGTB group (Table 3).
Delivery types among patients with complicated pregnancies. Table 4 shows the types of delivery in complicated pregnancies among the 3 groups. In the FMF group, there were 76 deliveries out of the 102 (74.5%) complicated pregnancies that were on time using natural methods. Thirteen (12.7%) deliveries required cesarean section. In the FGTB group, 36 out of 54 (66.7%) deliveries were on time and in natural ways. Delivery by cesarean section was performed in 4 (7.4%). In the control group, 74 out of 122 (60.7%) deliveries were on time and in the natural way. A cesarean section was performed in 16 cases (13.1%).
Perinatal outcome. Height-weight indices at birth were significantly lower in the FGTB group, compared with the FMF and control groups. The body weight of neonates was 1890 ± 790 g in the FGTB group, 2800 ± 500 g in the FMF group, and 2770 ± 580 g in the control group (P ≤ 0.05). The average height was 42 ± 7 cm in the second group compared with 48 ± 2.5 cm and 49 ± 3 cm in the first and third groups, respectively (P ≤ 0.001; data not shown).
Analysis of the FMF subgroups. Of the 211 patients with FMF, 139 had infertility. Infertile FMF patients had a significantly higher rate of attacks, earlier onset of FMF, and higher frequency of fever compared with fertile patients (Table 5). When the genotypes of fertile and infertile FMF patients were compared, it was found that the rate of infertile patients among the homozygotes was slightly higher than that of the compound heterozygotes but significantly higher than the rate of heterozygotes (Table 6). Further analysis revealed that infertility was significantly more common among patients homozygous for M694V and M680I. In addition, there was a clear correlation between the rate of infertility and delay in FMF diagnosis, irregular use of colchicine, or the use of low dose of the drug. Moreover, there was a higher rate of tuboperitoneal obstruction among FMF patients homozygous for MEFV mutations compared with the heterozygotes (Table 6).
DISCUSSION
In the present study, we compared the infertility causes and pregnancy outcome in 3 groups: FMF patients (group 1), FGTB (group 2), and healthy women with reproductive problems but without FMF or FGTB (group 3). Demographic data disclose that most patients in group 2 belonged to a low SES, whereas most patients in the control group were classified as middle SES. The FMF group included patients from both classes in almost equal numbers. This observation may explain the presence of tuberculosis (TB) infection in the FGTB group, since their hygiene and access to medical services were probably limited.
Our study shows that primary infertility was more common among FMF and FGTB patients, while secondary infertility was predominant in the control group. This observation is quite expected since the most common cause for infertility was tuboperitoneal obstruction. This complication occurred in FMF patients due to recurrent peritonitis and peritoneal adhesions, and in the FGTB group due to genital TB peritonitis. In the control group, the causes for infertility were either disovulation problems, endometriosis, or endometritis, following pelvic inflammatory disease. Zayed, et al reported that 18 out of 74 infertile women with FMF suffered from anovulation, whereas the majority, 56 (57.67%) patients, had excessive clear peritoneal fluid due to local inflammation19. These results are in accord with our observation. However, Nabil, et al claimed that the causes of infertility in patients with FMF are not different from those expected in the general population15. This observation is probably true in patients with FMF who are treated with colchicine, which can prevent the complications leading to infertility.
IVF was employed in the 3 studied groups. In the FMF and control groups, take-home baby indices were almost equal: 36.4% and 37.5%, respectively. Ozgur, et al reported that in their hands, the rate of take-home baby index was significantly higher, 58.3%20. Yilmaz, et al reported that successful pregnancy following IVF was achieved in only 3 out of 10 (30%) infertile FMF patients21. In the study of Zayed, et al, 26 (35%) out of 74 infertile women with FMF obtained successful pregnancy19. Thus, the IVF success rate in the last 2 studies resemble our results. The high success rate reported by Ozgur, et al was due to a higher number of treatment cycles, and the use of intra-cytoplasmic sperm injection when conventional IVF had failed20. The take-home baby index was much lower in the FGTB patients due to TB endometrial damage affecting successful implantation of the embryos.
The rate of successful deliveries was similar among all 3 groups (60%; Table 3). However, the rate of ectopic pregnancies was significantly higher in the FMF group. Frequent FMF attacks due to a lack of colchicine treatment can lead to strong uterine contractions, which may end up with ectopic implantation of the gestational sac.
The rate of early miscarriages and congenital malformation was quite similar in the FMF and control groups, whereas late miscarriages were more common in the FGTB group. Neonatal height and weight were also similar in groups 1 and 2, and significantly lower in the FGTB group. Nabil, et al reported a favorable pregnancy outcome in 26 patients with FMF treated with colchicine before and after pregnancy15. Their neonatal outcome was similar to that expected in the general population. Yasar, et al evaluated retrospectively, the outcome of pregnancy in 46 FMF patients and compared them with 138 control individuals16. They observed higher rates of cesarean delivery and low birth weight infants in the FMF group. However, rates of stillbirth did not differ between the groups. Preterm delivery rate was also higher in the FMF group, but this difference was not statistically significant16. In a population-based study, Ofir, et al compared the outcome of all pregnancies of women with and without FMF22. They found that FMF was an independent risk factor for preterm delivery. However, their perinatal outcome was comparable to the general population. Most of their FMF patients were treated with colchicine during pregnancy. In our study, higher rates of recurrent miscarriage occurred mainly in patients with FMF who were not on colchicine treatment. Thus, it seems that our results are in line with most observations of the above studies. However, the remaining differences may be explained by the different sizes of the studied groups, different study design, and different treatment regime with colchicine.
Comparing fertile and infertile FMF patients disclosed that infertility was clearly associated with carriage of M694V or M680I mutations (homozygotes). These genotypes are associated with more severe disease as early onset of the disease, more frequent attacks, and more joint involvement. In addition, infertility and bad pregnancy outcome were more common among females who did not take colchicine or were treated inadequately.
Many FMF patients are afraid to take colchicine during pregnancy due to the theoretical teratogenic potential of the drug. Ben-Chetrit, et al reported no difference in early abortions, late abortions, and congenital malformations between FMF patients who took colchicine during pregnancy and health-, age- and ethnicity-matched controls12. In a study by Diav-Citrin, et al, 238 colchicine-exposed pregnancies were compared with 964 pregnancies without colchicine exposure23. The results showed again that colchicine use did not cause increased cytogenetic risk.
The major drawbacks of this study are those inherent in all retrospective studies. However, the relatively large size of the groups studied and the large amount of data recruited may further strengthen our conclusions.
In summary, our study results show that the FMF group resembles the FGTB group regarding the etiologies for infertility. However, regarding IVF and pregnancy outcomes, the FMF group more resembles the control group. We observed that FMF has no significant effect on the frequency of early or late abortions, congenital malformations, or late obstetric complications. The slight predominance of early miscarriages and preterm delivery reported by others may be explained by a lack of adequate colchicine treatment. Low SES may also have a similar effect due to a limited access to colchicine and good healthcare. In a recent study, Atas, et al show that FMF disease onset (< 20 years), disease severity, and colchicine nonresponse were independent risk factors for FMF-associated infertility24. This observation is in line with our findings and supports the recommendation that patients with FMF should start colchicine immediately at diagnosis and continue treatment even during pregnancy, in order to control the disease and prevent its potential obstetric complications.
Footnotes
The authors declare no conflicts of interest.
- Accepted for publication September 9, 2020.
- Copyright © 2021 by the Journal of Rheumatology