SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Objective

Administration of nonsteroidal antiinflammatory drugs (NSAIDs) may impair fertility. The occurrence of the luteinized unruptured follicle (LUF) syndrome was assessed in women with inflammatory arthropathies exposed to NSAIDs and compared to that in nonexposed women.

Methods

Fourteen patients with inflammatory rheumatic disease, 29 women with noninflammatory musculoskeletal conditions, and 449 women not exposed to NSAIDs were studied by intravaginal ultrasound monitoring for follicular development and ovulation in 1 or more menstrual cycles. Disease activity was assessed in inflammatory rheumatic disease.

Results

In 59 monitored cycles of patients with continuous NSAID exposure, 35.6% of LUF syndromes occurred compared to 3.4% of LUF syndromes in untreated women (P < 0.001). Etoricoxib was responsible for 75% of LUF syndromes in patients exposed continuously, whereas diclofenac generated 15% of LUF syndromes. An ibuprofen dosage of 1,600 mg/day did not induce LUF syndrome either at continuous periovulatory or discontinuous exposure. Interestingly, the frequency of LUF syndrome was 46.2% in patients with inactive inflammatory disease compared to 15% in patients with active disease (P = 0.023). Etoricoxib generated LUF syndrome in 94.2% of the cases with inactive disease versus 28.6% in patients with active disease (P = 0.003).

Conclusion

NSAIDs increased the risk of the LUF syndrome, particularly in patients with inactive disease. The selective cyclooxygenase 2 (COX-2) inhibitor etoricoxib was a more potent inductor of LUF syndrome than nonselective COX inhibitors. Continuous periovulatory exposure to NSAIDs should be avoided when planning a pregnancy in patients with rheumatic diseases.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Patients with rheumatic diseases receiving nonsteroidal antiinflammatory drugs (NSAIDs) intermittently or long term have a risk of developing the luteinized unruptured follicle (LUF) syndrome, resulting in secondary infertility. The LUF syndrome can occur spontaneously (1, 2), but exposure to NSAIDs that are inhibitors of prostaglandins (PG) increases its occurrence both in animals and humans (1–3). The preovulatory release of PG is necessary to initiate a cascade of catalytic reactions, resulting in the rupture of the follicle wall and release of the oocyte (3, 4).

Previous studies comprising series of healthy women and case reports of women with rheumatic diseases (1, 5–11) demonstrated the occurrence of the LUF syndrome with exposure to inhibitors of cyclooxygenase 1 (COX-1) and COX-2. Placebo-controlled studies of healthy women exposed to NSAIDs showed delayed rupture of the follicle, but no modification of the hormonal cyclicity, although a decrease of progesterone was observed in some studies (5, 7). In the rheumatology literature, 11 patients infertile during the use of NSAIDs have been reported (8–11). Comparisons to women not exposed to NSAIDs were not made. Only 3 of these patients had ultrasound monitoring of ovulation (8). The other 8 women were infertile and were suspected to have LUF syndrome because conception occurred after discontinuation of NSAIDs (9–11). Drugs involved were inhibitors of COX-1 and -2: indomethacin, naproxen, diclofenac, and piroxicam.

The present study was performed to compare the occurrence of LUF syndrome in patients with rheumatic diseases exposed to NSAIDs to the occurrence of LUF syndrome in nonexposed women. Secondary objectives were to identify the influence of different types of COX inhibitors and exposure regimens on the occurrence of the LUF syndrome. A possible influence of disease activity was also investigated.

PATIENTS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

This prospective study was approved by the Ethics Committee of the University of Medicine and Pharmacy, Cluj Napoca, and was performed between 2007 and 2009 at the County Hospital Turda/Clinical Rehabilitation, Hospital Cluj, Division of Rheumatology, and Gynecology Clinic I, Cluj. Included after written consent were women of fertile age, with at least 1 pregnancy in the past, who were divided in 3 groups. Group 1 included 14 patients (10 with rheumatoid arthritis [RA], 4 with ankylosing spondylitis [AS]), and 2 control groups: group 2 consisted of 29 healthy women exposed to NSAIDs for noninflammatory conditions and group 3 included 449 women from the same area not exposed to NSAIDs, examined because of secondary infertility.

Ultrasound monitoring.

The occurrence of ovulation in multiple menstrual cycles was monitored by intravaginal ultrasound (12). From day 10 of the menstrual cycle, the dominant follicle was monitored up to 48 hours after the urinary detection of the luteinizing hormone (LH) surge. The occurrence of ovulation was estimated to be in the interval of 36 ± 4 hours after the LH surge (urinary stick Clearblue Easy Ovulation Tests; SPD Swiss Precision Diagnostics) and identified by a reduction of 50% or more of the maximum dominant follicle diameter, sometimes accompanied by a small Douglas sack effusion (12). There were no significant differences in the diameter of the dominant follicle or the length of the menstrual cycle in the 3 groups of women studied. The persistence of the unruptured follicle for more than 48 hours apart from the LH surge was diagnosed as LUF syndrome. Repeated ultrasound examinations and blood tests were performed at precise hour intervals after the LH peak detection in each menstrual cycle.

NSAID exposure regimens were defined as: 1) continuous periovulatory regimen: daily exposure starting with day 1 of the menstrual cycle up to 48 hours after the LH surge, 2) discontinuous regimen: daily exposure starting with day 1 and discontinuing after day 8 of the menstrual cycle, and 3) no NSAIDs. Dosages of NSAIDs used were: etoricoxib 90 mg/day as a single dose, diclofenac 3 × 50 mg/day, ketoprofen 2 × 100 mg/day, ibuprofen 4 × 400 mg/day, celecoxib 2 × 100 mg/day, and nimesulide 2 × 100 mg/day. The same NSAID and dose were kept throughout each menstrual cycle. Patients who did not strictly follow the investigation protocol or medication were eliminated from the study. All patients in group 2 received a continuous periovulatory regimen because of a prolonged noninflammatory musculoskeletal condition.

In group 1, assessment of disease activity was performed in every menstrual cycle at day 8 and at 72 hours after the LH surge using the Disease Activity Score in 28 joints (DAS28) for patients with RA (13) and the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) for patients with AS (14). Laboratory analysis was performed between days 8 and 10 of the menstrual cycle, with measurements of erythrocyte sedimentation rate and C-reactive protein (CRP) level. Inactive disease was defined as a CRP level of <0.6 mg/dl, a DAS28 of <2.6 in RA, and a BASDAI score of <2 in AS. All medications used by the patients were recorded.

Statistical analysis.

Chi-square test and Fisher's exact test were used for the statistical analysis. A P value of 0.05 was considered statistically significant.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

The 14 patients in group 1 (mean ± SD age 33.5 ± 9.1 years) were monitored during 59 cycles with the continuous periovulatory regimen, 20 cycles occurring during active disease, and 39 cycles during inactive disease (Tables 1 and 2). All patients with AS ingested only NSAIDs. The patients with RA had the following accompanying medications: salazopyrine 2 gm/day in patients 1, 5, 13, and 14; hydroxychloroquine 2 × 200 mg/day in patients 6, 7, 9, 10, and 11; and azathioprine 2.5 mg/kg/day in patient 8.

Table 1. Influence of disease activity and type of NSAID on the occurrence of LUF syndrome in 10 women with RA and 4 women with AS exposed to a continuous periovulatory regimen*
DiagnosisActive diseaseInactive disease
No. of cycles monitoredNSAID usedNo. of LUF syndromes developedNo. of cycles monitoredNSAID usedNo. of LUF syndromes developed
  • *

    NSAID = nonsteroidal antiinflammatory drug; LUF = luteinized unruptured follicle; RA = rheumatoid arthritis; AS = ankylosing spondylitis.

RA11 etoricoxib071 etoricoxib, 3 ibuprofen (3 without NSAID exposure)1 etoricoxib
AS4None222 etoricoxib2 etoricoxib
AS52 celecoxib, 3 ketoprofen0
AS66 diclofenac0
RA22 etoricoxib1 etoricoxib61 etoricoxib, 2 ibuprofen, 1 diclofenac (2 without NSAID exposure)1 etoricoxib, 1 diclofenac
RA51 etoricoxib, 2 diclofenac, 2 ibuprofen1 etoricoxib, 1 diclofenac
RA52 etoricoxib, 3 ibuprofen2 etoricoxib
RA22 etoricoxib022 etoricoxib2 etoricoxib
RA52 etoricoxib (3 without NSAID exposure)2 etoricoxib
RA42 etoricoxib, 1 ibuprofen2 etoricoxib
RA53 diclofenac, 2 ibuprofen1 diclofenac
AS31 etoricoxib, 1 diclofenac, 1 ibuprofen011 etoricoxib1 etoricoxib
RA21 etoricoxib, 1 diclofenac1 etoricoxib33 diclofenac0
RA22 diclofenac042 etoricoxib, 1 diclofenac, 1 ibuprofen2 etoricoxib
Table 2. Effect of continuous exposure to different NSAIDs on the occurrence of LUF syndrome in 14 patients with inflammatory rheumatic disease and 29 women with chronic musculoskeletal disease*
Continuous exposure to NSAIDsNo. of cycles exposed to NSAIDs in group 1 (active/inactive disease)No. of cycles exposed to NSAIDs in group 2Development of LUF syndrome in group 1Development of LUF syndrome in group 1/group 2 (%)
Active diseaseInactive disease
  • *

    NSAIDs = nonsteroidal antiinflammatory drugs; LUF = luteinized unruptured follicle.

Etoricoxib24 (7/17)921618/5 (75/55.6)
Diclofenac20 (7/13)14123/2 (15/14.5)
Ketoprofen3 (3/0)3000
Ibuprofen10 (1/9)0000
Celecoxib2 (2/0)0000
Nimesulide03000

Table 1 shows NSAIDs used in each of the 14 patients, the number of cycles monitored, and the number of LUF syndromes occurring. Changes of NSAIDs in individual patients during the study period were either due to insufficient effect of the former medication or wish for an NSAID with a shorter half-life than etoricoxib. Five patients in group 1 changed after the initial continuous regimen to the discontinuous or no NSAIDs exposure regimen (5 cycles with ibuprofen, 12 with no exposure) because of pregnancy desire. No LUF syndrome occurred in these 17 cycles.

The 29 patients in group 2 (mean ± SD age 34.5 ± 3.4 years) had continuous periovulatory exposure to NSAIDs. Table 2 shows the NSAIDs administered for a prolonged noninflammatory musculoskeletal condition (low back pain). The LUF syndrome occurred in 7 (24.1%) of 29 monitored cycles (Table 2).

Group 3 comprising 449 women consulting for secondary infertility (mean ± SD age 28.1 ± 1.2 years) was not exposed to NSAIDs. The LUF syndrome occurred in 59 (3.4%) of 1,744 monitored cycles.

When analyzing the occurrence of LUF syndrome in the 3 groups of women by the chi-square test and Fisher's exact test, LUF syndrome occurred significantly more often in women exposed to NSAIDs in a continuous periovulatory regimen than in women not exposed to NSAIDs (group 1 versus group 3: P < 0.001 and group 2 versus group 3: P < 0.001). There was no significant difference in the occurrence of LUF syndrome between patient groups (group 1 versus group 2: P = 0.336), regardless of the type of NSAID administered. Among NSAIDs, LUF syndrome occurred most frequently during the use of etoricoxib followed by diclofenac, whereas no LUF syndrome occurred in any patient group during exposure to ibuprofen, ketoprofen, celecoxib, or nimesulide, regardless of exposure regimen or active or inactive disease (Table 2).

In group 1, the incidence of LUF syndrome was significantly higher at continuous periovulatory exposure to NSAIDs in inactive disease (46.2%) compared to active disease (15%; P = 0.023). Etoricoxib generated LUF syndrome in 94.2% of the cases with inactive disease versus 28.6% in patients with active disease (P = 0.003) (Table 1).

Four pregnancies occurred in cycles without exposure to NSAIDs and 1 on a discontinuous regimen with ibuprofen. All pregnant women had at least 1 drug-induced LUF syndrome detected earlier during the study.

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

Our study including the largest group of patients and controls monitored by ultrasound confirmed the significantly increased occurrence of LUF syndrome in women taking NSAIDs compared to women not treated with NSAIDs. In contrast to previous case reports of patients with rheumatic disease, our study investigated different types of COX inhibitors as well as exposure to different NSAIDs in the same patient in subsequent cycles, and this allowed us to observe drug-related differences in LUF syndrome expression. The apparent differences in induction of LUF syndrome between various NSAIDs are of great interest, since the use of NSAIDs has recently been studied for monthly or emergency contraception (15, 16).

NSAIDs with a long half-life and preferential COX-2 inhibition such as etoricoxib did inhibit follicle rupture more often than nonselective NSAIDs or NSAIDs with a short half-life. The enzymes involved in PG synthesis are the 2 isoforms of COX-1 and COX-2. For the process of ovulation, COX-2 seems to be of greater importance than COX-1 (3, 4). This may explain why etoricoxib, a selective COX-2 inhibitor, induced the LUF syndrome much more frequently than the other NSAIDs used by our patients. Furthermore, nonselective NSAIDs and those with a shorter half-life than etoricoxib may allow proper COX production in drug-free intervals. Another explanation could be dose dependency of inhibition of follicle rupture. A dose dependency of ovulation inhibition was shown in a double-blind crossover study of 22 healthy women exposed to either 15 mg/day or 30 mg/day of meloxicam in the late follicular phase (16). Dysfunctional ovulation was observed in 11 (50%) of 22 cycles treated with 15 mg/day and 20 (90.9%) of 22 cycles treated with 30 mg/day of meloxicam (16). Complete inhibition of ovulation has been achieved by 150–200 mg of indomethacin (5, 9). This indicates that prostanoids must be reduced below a certain threshold to result in LUF syndrome. The full antiinflammatory dose of etoricoxib is given in a single dose, whereas diclofenac, ketoprofen, and ibuprofen are given in divided doses. Patients who need to take a drug several times a day are often not compliant and take less than was prescribed. Contrary to the study by Uhler et al (17), where 3 of 5 healthy women exposed to 2,400 mg of ibuprofen daily showed a delay of follicle rupture, none of our patients developed LUF syndrome on 1,600 mg/day of ibuprofen. This indicates that our patients were not exposed to the full antiinflammatory dose of ibuprofen necessary for induction of the LUF syndrome.

The overall rate of LUF syndrome in our patients with inflammatory rheumatic disease using NSAIDs approximated 35%, less than the rates of approximately 75–100% observed in healthy women exposed to full doses of COX inhibitors (5, 15, 17). Surprisingly, patients with inactive disease and women with noninflammatory conditions exposed to a continuous regimen of NSAIDs showed a similar and much higher occurrence of the LUF syndrome than RA and AS patients with active disease. To our knowledge, this observation has not been reported before. A possible explanation could be the fact that patients with active disease need a higher level of COX-2 inhibition to suppress the general inflammation. Expression of COX-2 is increased at inflammation sites with subsequent preferential binding of the NSAIDs at these sites (18), resulting in reduced levels in the follicular fluid. This would allow the release of proper follicular PG concentrations, followed by adequate timing of follicle rupture and release of the mature oocyte.

Short-term use of NSAIDs during the first 8 days of the menstrual cycle was not associated with LUF syndrome compared to continuous use extending to midcycle. This confirms the observation of Smith et al, who demonstrated normal ovulation in their 3 patients with previous LUF syndrome when naproxen, piroxicam, and diclofenac were discontinued at day 8 or 10 of the menstrual cycle (8). It also emphasizes the importance of timing when administering NSAIDs to cycling women. Successful ovulation inhibition requires administration of NSAIDs during the late follicular phase, starting when the leading follicle approaches a diameter of 18 mm (15, 16).

The limits of our study are the limited number of patients with inflammatory disease (14 with 76 monitored cycles), the small number of cycles with the discontinuous exposure regimen, and the fact that the women in group 3 were not the “general population” but presented with infertility after a previous pregnancy. One may anticipate that the prevalence of the LUF syndrome in a nonselected healthy population is even lower than was found in group 3 of our study. Five patients were lost for further monitoring of cycles because of conception, which also confirms that NSAIDs induce reversible infertility.

To conclude, periovulatory exposure to the full antiinflammatory dose of NSAIDs increases the risk for development of the LUF syndrome, particularly in patients with inactive rheumatic disease. Highly selective COX-2 inhibitors with a long plasma half-life seem to have the most negative impact on ovulation. Therefore, female patients of fertile age in need of NSAIDs should be treated with the lowest effective dose and drugs with a short half-life. Discontinuing or replacing NSAIDs with analgesic drugs starting with day 8 of the menstrual cycle may represent a feasible alternative for women with rheumatic disease who want to become pregnant, since NSAIDs do not permanently reduce fertility.

AUTHOR CONTRIBUTIONS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Mihaela C. Micu had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. Mihaela C. Micu, Romeo Micu, Ostensen.

Acquisition of data. Mihaela C. Micu, Romeo Micu.

Analysis and interpretation of data. Mihaela C. Micu, Romeo Micu, Ostensen.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. PATIENTS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. AUTHOR CONTRIBUTIONS
  8. REFERENCES