- Email: [email protected]
Short cervix on ultrasound: Does indomethacin prevent preterm birth?
American Journal of Obstetrics and Gynecology (2006) 195, 809–13
www.ajog.org
Short cervix on ultrasound: Does indomethacin prevent preterm birth? Vincenzo Berghella, MD,a Orion A. Rust, MD,b Sietske M. Althuisius, MDc Departments of Obstetrics and Gynecology, Division of Maternal-Fetal Medicine,a Jefferson Medical College of Thomas Jefferson University, Philadelphia, PA; Division of Maternal-Fetal Medicine,b Lehigh Valley Hospital and Health Network, Allentown, PA; VU Medical Center,c Amsterdam, The Netherlands Received for publication March 9, 2006; revised May 8, 2006; accepted June 1, 2006
KEY WORDS Preterm birth Cervical length Ultrasound
Objective: Our aim was to estimate if indomethacin therapy prevents preterm birth (PTB) in women with a short cervical length (CL) on transvaginal ultrasound (TVU). Study design: Individual-level data from all randomized trials including asymptomatic women with a short CL on TVU were analyzed for use of indomethacin at the time of the short CL. The trials eligible would be ones that randomized women with a short CL !25 mm, identified between 14 and 27 weeks. The eligible trials randomized such women to receive either cerclage or no cerclage. Only women who did NOT receive cerclage were analyzed. Exclusion criteria were major fetal anomaly and cerclage. We compared demographics, risk factors, and outcomes in women who, at the time of the short CL, received indomethacin or not. Primary outcome was PTB !35 weeks. Results: Three of the 4 randomized trials identified had databases which recorded indomethacin use at the time of the short CL. A total of 139 women with a short CL !25 mm identified at 14 to 27 weeks were identified. Of these women, 99 (71.2%) received indomethacin, and 40 (28.8%) did not. Demographics and risk factors, including previous PTB (45.5% vs 62.5%; P = .11), were similar in the 2 groups. The primary outcome of PTB !35 weeks occurred in 29.3% (29/99) of women who received indomethacin, and 42.5% (17/40) of women who did not receive indomethacin (RR 0.69, 95% CI 0.44-1.13). PTB !24 weeks occurred in 1.0% (1/99) versus 7.5% (3/40), respectively (RR 0.14; 95% CI 0.02-0.92). Incidence of perinatal death was similar in the 2 groups (6% vs 10%; RR 0.61, 95% CI 0.19-1.95). Conclusion: Indomethacin therapy for asymptomatic women who have a short CL !25 mm on TVU at 14 to 27 weeks and do not receive a cerclage did not prevent spontaneous PTB !35 weeks, but did prevent PTB !24 weeks. Further research including larger numbers and a randomized trial design is necessary to further clarify the effectiveness as well as the risks of this therapy. Ó 2006 Mosby, Inc. All rights reserved.
Preterm birth is the most common cause of perinatal morbidity and mortality. A short cervical length on transvaginal ultrasound is one of the best risk-
assessment methods for prediction of preterm birth.1-3 The appropriate intervention after identifying a woman as high-risk for spontaneous preterm birth because of a
Presented at the Twenty-Sixth Annual Meeting of the Society for Maternal Fetal Medicine, Miami, FL, January 30-February 4, 2006. Reprints not available from the authors. 0002-9378/$ - see front matter Ó 2006 Mosby, Inc. All rights reserved. doi:10.1016/j.ajog.2006.06.007
810 Table I
Berghella, Rust, and Althuisius Descriptive data for each trial
Study
Population screened
Rust6
All singletons, twins, and triplets Singletons with suspected cervical incompetence; twins All singletons Singletons with risk factors for PTB and twins
Althuisius7 To8 Berghella9
GA at screening (wk)
Cervical length (mm)
Funneling (%)
16-24
!25
R 25
14-27
!25
–
22-24 14-24
%15 !25
– O25
PTB, Preterm birth; GA, gestational age; N/A, not available.
sonographic short cervix remains to be determined. Several studies have shown that asymptomatic women with a short cervix on ultrasound are having uterine contractions.4,5 Indomethacin is a nonselective, nonsteroid, anti-inflammatory agent that inhibits the transformation of arachidonic acid into prostaglandins. Prostaglandins are important not only in uterine contractility, but also in the induction of cervical ripening. Cerclage has been so far the only intervention studied to prevent preterm birth once a short cervical length has been detected in asymptomatic women.6-9 Only 1 of these randomized trials found a significant decrease in preterm birth before 34 weeks in favor of the cerclage group.7 In that trial all women treated with a cerclage received indomethacin, while the women not treated with a cerclage did not. This implicates that the indomethacin might have contributed to the positive outcome in the cerclage group. The objective of this study is to systematically review all randomized trials of asymptomatic women with a short cervical length on second-trimester transvaginal ultrasound, examining the effect of indomethacin on prevention of preterm birth.
Material and methods All randomized trials including women with a short cervical length were searched in MEDLINE, PUBMED, EMBASE, and Cochrane Library from 1966 until 2004, inclusive. All manuscripts were reviewed for pertinent references. Selection included asymptomatic pregnant women, found by second-trimester transvaginal ultrasound to have a short cervical length. Exclusion criteria were major fetal anomaly, twin gestation, and cerclage. Only randomized study design trials were included. The eligible trials identified randomized such women to receive either cerclage or no cerclage. Only women who did NOT receive cerclage were analyzed in this study. We compared demographics, risk factors, and outcomes in women who, at the time of the short cervical length, received indomethacin or not. Use of cerclage, tocolytics, or any other intervention around the time of short cervical length was assessed. A metaanalysis on the effect of cerclage on incidence of preterm
birth has already been published.10 Given that characteristics and outcomes reported differed among the manuscripts, primary authors of each trial selected for inclusion were asked for the actual databases, to allow for a patient-level meta-analysis. Primary outcome was PTB !35 weeks. Data analysis was accomplished by relative risks, with 95% confidence intervals. Secondary outcomes reported include incidence of preterm birth at different gestational age cut-offs, incidences of preterm premature rupture of membranes (PPROM), and of perinatal mortality (defined as the sum of intrauterine fetal demise R20 weeks and neonatal demise). Statistical analysis included t test for continuous variables and chi-square or Fisher exact test for categorical variables, and was performed using SPSS, version 12 (Chicago, IL).
Results Four trials6-9 met the inclusion criteria of intervention for short cervical length on second-trimester transvaginal ultrasound (Table I). All primary authors agreed to provide us with the actual databases, including all women randomized. Two authors6,7 supplied data on more randomized women than in their original publications. Three trials recorded use of indomethacin at the time of the short cervical length,6,7,9 and 1 did not.8 The 3 trials included 354 women who had a short cervical length between 14 and 27 weeks. After excluding women who received a ultrasound-indicated cerclage (n = 177), women with multiple gestations (n = 26), and those who received a physical exam-indicated (rescue) cerclage (n = 12), a total of 139 women with singleton gestation, a short CL !25 mm identified at 14 to 27 weeks and no cerclage were identified and analyzed. Of these women, 99 (71.2%) received indomethacin, and 40 (28.8%) did not. All the women in the indomethacin group received indomethacin 100 mg per rectum, followed by 50 mg orally every 6 hours for 48 to 72 hours right after the short CL !25 mm was identified. The majority of women in the indomethacin group came from 1 study,6 and the majority of no indomethacin women from another.9 Demographics and risk factors, including previous PTB, were similar in the
Berghella, Rust, and Althuisius Table I
811
(Continued)
Study
Total sample size (n)
Pericerclage tocolytics (indomethacin)
Rust6
241
Both groups
Althuisius7
52
To8 Berghella9
253 61
Cerclage group only Per obstetrician Per obstetrician
Singleton with no cerclage and indomethacin information (n)
Indomethacin dose Indomethacin 100 mg per rectum, followed by 50 mg orally q6 hours for 48-72 hours Indomethacin not given to no cerclage group N/A Indomethacin 100 mg per rectum, followed by 50 mg orally q6 hours for 48 hours
96 (indomethacin) 14 (no indomethacin) N/A 29 (3 indomethacin/26 no indomethacin)
PTB, Preterm birth; GA, gestational age; N/A, not available.
Table II
Demographics and risk factors
Previous PTB 16-36 wk Cone O1 D&E Short CL (mm)
Table III
Outcomes
Indomethacin (n = 99) (%)
No indomethacin (n = 40) (%)
P value
47 (47)
25 (63)
.11
PTB !35 wk
29 (29)
17 (43)
24 (24) 20 (20) 17.5 G6.6
4 (10) 13 (33) 18.0 G6.8
.06 .12 .70
PTB !32 wk
20 (20)
14 (35)
PTB !24 wk
1 (1)
3 (8)
10 (10)
15 (38)
6 (6)
4 (10)
PTB, Preterm birth; D&E, dilatation and evacuation; CL, cervical length.
2 groups (Table II). The primary outcome of PTB !35 weeks occurred in 29% of women who received indomethacin, and 43% of women who did not receive indomethacin (RR 0.69, 95% CI 0.44-1.13). Incidences of PTB !24 weeks and PPROM were decreased in the indomethacin compared to the control group, while perinatal death was similar in the 2 groups (Table III). In the subgroup of women with previous preterm birth, preterm birth !35 weeks occurred in 34% (16/47) of women who received indomethacin, and 48% (12/25) of women who did not receive indomethacin (RR 0.71, 95% CI 0.42-1.28).
Comment Indomethacin therapy for asymptomatic women who have a short CL !25 mm on TVU at 14 to 27 weeks and do not receive a cerclage was not associated with a significant effect on PTB !35 weeks, but was associated with prevention of both preterm birth !24 weeks and of PPROM. Perinatal mortality was also not significantly different, with a trend toward benefit. For the trend in decrease in preterm birth !35 weeks to reach statistical significance, 200 women with similar characteristics in each group would be needed, assuming an alpha of 0.05 and beta of 0.8. Further research is necessary to clarify the effectiveness as well as the risks of this therapy. There are no other studies on the effect of indomethacin for prevention of preterm birth in asymptomatic women with a short cervical length. In this patient population, the only intervention studied so far has
Indomethacin No indomethacin RR (n = 99) (%) (n = 40) (%) (95% CI)
PPROM Perinatal death
0.69 (0.44-1.13) 0.58 (0.33-1.04) 0.14 (0.02-0.92) 0.27 (0.13-0.54) 0.61 (0.19-1.95)
PTB, Preterm birth; PPROM, premature preterm rupture of membranes; RR, relative risk; CI, confidence interval.
been cervical cerclage. A recent meta-analysis demonstrated prevention of preterm birth in singleton gestations with a short cervical length if cerclage was placed, especially in such women who also had a previous preterm birth.10 In women with a short cervical length but instead carrying twin gestations, cerclage was associated with an increase in preterm birth compared to no cerclage. The results of this meta-analysis should be confirmed by a randomized study. Women who received cerclage for the short cervical length (ultrasound-indicated cerclage) were excluded from our study, to avoid this counfounder. There is insufficient evidence to assess if women who receive cerclage for short cervical length benefit from indomethacin therapy. Almost all women randomized to cerclage in the 4 trials6-9 published so far also received indomethacin, so that no control group (cerclage but no indomethacin) was available for comparison. No studies have been reported on this clinical scenario. Women screened by scheduled transvaginal ultrasound report no symptoms before being identified to have a short cervical length before 24 weeks. The etiology of the short cervical length has been postulated to be either cervical insufficiency,11 infection,12 contractions,4,5 or a combination of these or other factors. Only women with some cervical insufficiency component responsible for the short cervical length might benefit
812 from cerclage. When placed on the tocomonitor after the short cervix has been detected, contractions can easily and frequently be detected in the majority of these asymptomatic women.4,5 This association between short cervix and uterine contractions does not clarify yet if the contractions ‘cause’ the short cervix, or vice versa, or if these changes both derive from a common factor. The fact that contractions are present, though, could make the intervention of tocolysis hypothetically and potentially advantageous. Indomethacin has been studied extensively as a tocolytic in symptomatic women. Compared to placebo, indomethacin is associated with a significant reduction in preterm birth !37 weeks as well as a trend towards a reduction in delivery within 48 hours of initiation of treatment.13 Compared with betamimetics or magnesium, cyclo-oxygenase (COX) inhibitors such as indomethacin are associated with a significant reduction in the number of women delivering within 48 hours and in the incidence of preterm birth !37 weeks.13 In all trials, there was only 1 case of antenatal closure of the ductus arteriosus in 403 women who received short-term tocolysis (up to 48 hours) with COX inhibitors (mainly indomethacin). There was no increase in the incidence of patent ductus arteriosus postnatally (8 treated with COX inhibitors vs 8 treated with placebo or other tocolytics).13 No difference in incidences of intraventricular hemorrhage, bronchopulmonary dysplasia, patent ductus arteriosus, necrotizing enterocolitis, or perinatal mortality was noted in a review aimed at evaluating safety of 1621 fetuses exposed to indomethacin in trials and observational studies compared to controls.14 It would seem reasonable that a short course (48 hours) of indomethacin in the second trimester, before 24 weeks, would not be associated with severe fetal or neonatal complications, but future research should closely assess the safety and risks of this intervention. Strengths of our study are the novel intervention in this population of women at risk for preterm birth because of a short cervical length while asymptomatic. Data analyzed were from prospective, randomized studies, and represent the only assessment of this therapy for this indication thus far in the literature. A shortcoming might have been the sample size, which, if larger, might have lead to different results (eg, significance for preterm birth at !32 weeks). The fact that criteria for administration were either based on trial design6,7 or per obstetrician9 may be another limitation. Another limitation is the fact that long-term infant outcome could not be obtained. The fact that perinatal mortality was nonsignificantly decreased by about 39% in the indomethacin group, with only 1 such outcome in this group, is reassuring. This study shows that indomethacin is associated with a trend toward statistical significance with a 31%
Berghella, Rust, and Althuisius decrease in preterm birth !35 weeks, and with a significant 86% decrease in preterm birth !24 weeks in women with a short cervix before 24 weeks. These results should serve as pilot data for future placebocontrolled trials to assess if indomethacin is beneficial in preventing preterm birth in asymptomatic women with short cervical length on transvaginal ultrasound and either no cerclage or a cerclage in place. Other interventions such as antibiotics, progesterone, bed rest, etc, should also be studied in the future with proper trials. Trials addressing the efficacy of indomethacin should include a specific asymptomatic population (eg, women with previous preterm birth), accurately define short CL (eg, !25 mm), and gestational age (eg, 14-23 6/7 weeks, or even later), dose and duration of therapy (eg, 50 mg every 6 hours for 48 hours), and outcome (eg, preterm birth !35 weeks). Probably this trial is justified both in women with or those without a history-indicated cerclage in place. Other interventions such as those mentioned above should be standardized, and/or controlled for at randomization. Even if indomethacin with or without cerclage is confirmed to benefit certain populations of pregnant women, the incidence of preterm birth with a short cervical length remains high. Other interventions or combinations of interventions should also be tested in the future with properly designed trials.
References 1. Iams JD, Goldenberg RL, Meis PJ, Mercer BM, Moawad A, Das A, et al. The length of the cervix and the risk of spontaneous premature delivery. N Engl J Med 1996;334:567-72. 2. Berghella V, Daly SF, Tolosa JE, DiVito MM, Chalmers R, Garg N, et al. Prediction of preterm delivery with transvaginal ultrasonography of the cervix in patients with high-risk pregnancies: does cerclage prevent prematurity. Am J Obstet Gynecol 1999; 181:809-15. 3. Owen J, Yost N, Berghella V, Thom E, Swain M, Dildy GA, et al. Mid-trimester endovaginal sonography in women at high risk for spontaneous preterm birth. JAMA 2001;286:1340-8. 4. Berghella V, Iams JD, Newman RB, MacPherson C, Goldenberg RL, Mueller-Heubach E, et al. Frequency of uterine contractions in asymptomatic pregnant women with or without a short cervix on transvaginal ultrasound scan. Am J Obstet Gynecol 2004;191: 1253-6. 5. Lewis D, Pelham JJ, Done E, Sawhney H, Talucci M, Berghella V. Uterine contractions in asymptomatic women with a short cervix on ultrasound. J Matern Fetal Neonatal Med 2005;18: 325-8. 6. Rust OA, Atlas RO, Reed J, van Gaalen J, Balducci J. Revisiting the short cervix detected by transvaginal ultrasound in the second trimester: why cerclage therapy may not help. Am J Obstet Gynecol 2001;185:1098-105. 7. Althuisius SM, Dekker GA, Hummel P, Bekedam DJ, van Geijn HP. Final results of the cervical incompetence prevention randomized cerclage trial (CIPRACT): therapeutic cerclage with bed rest versus bed rest alone. Am J Obstet Gynecol 2001;185: 1106-12.
Berghella, Rust, and Althuisius 8. To MS, Alfirevic Z, Heath VCF, Cacho AM, Williamson PR, Nicolaides KH. Cervical cerclage for prevention of preterm delivery in women with short cervix: randomized controlled trial. Lancet 2004;363:1849-53. 9. Berghella V, Odibo AO, Tolosa JE. Cerclage for prevention of preterm birth in women with a short cervix found on transvaginal ultrasound examination: a randomized trial. Am J Obstet Gynecol 2004;191:1311-7. 10. Berghella V, Obido AO, To MS, Rust OA, Althiusius SM. Cerclage for short cervix on ultrasound: meta-analysis of trials using individual patient-level data. Obstet Gynecol 2005;106:181-9. 11. Iams JD, Johnson FF, Sonek J, Sachs L, Gebauer C, Samuels P. Cervical competence as a continuum: a study of ultrasonographic
813 cervical length and obstetric performance. Am J Obstet Gynecol 1995;172:1097-106. 12. Guzman ER, Shen-Schwarz S, Benito C, Vintzileos AM, Lake M, Lai YL. The relationship between placental histology and cervical ultrasongraphy in women at risk for pregnancy loss and spontaneous preterm birth. Am J Obstet Gynecol 1999; 181:793-7. 13. King J, Flenady V, Cole S, Thornton S. Cyclo-oxygenase (COX) inhibitors for treating preterm labour. Cochrane Database of Systematic Reviews 3; 2005. 14. Loe SM, Sanchez-Ramos L, Kaunitz A. Assessing the neonatal safety of indomethacin tocolysis: a systematic review with metaanalysis. Obstet Gynecol 2005;106:173-9.
www.ajog.org
Short cervix on ultrasound: Does indomethacin prevent preterm birth? Vincenzo Berghella, MD,a Orion A. Rust, MD,b Sietske M. Althuisius, MDc Departments of Obstetrics and Gynecology, Division of Maternal-Fetal Medicine,a Jefferson Medical College of Thomas Jefferson University, Philadelphia, PA; Division of Maternal-Fetal Medicine,b Lehigh Valley Hospital and Health Network, Allentown, PA; VU Medical Center,c Amsterdam, The Netherlands Received for publication March 9, 2006; revised May 8, 2006; accepted June 1, 2006
KEY WORDS Preterm birth Cervical length Ultrasound
Objective: Our aim was to estimate if indomethacin therapy prevents preterm birth (PTB) in women with a short cervical length (CL) on transvaginal ultrasound (TVU). Study design: Individual-level data from all randomized trials including asymptomatic women with a short CL on TVU were analyzed for use of indomethacin at the time of the short CL. The trials eligible would be ones that randomized women with a short CL !25 mm, identified between 14 and 27 weeks. The eligible trials randomized such women to receive either cerclage or no cerclage. Only women who did NOT receive cerclage were analyzed. Exclusion criteria were major fetal anomaly and cerclage. We compared demographics, risk factors, and outcomes in women who, at the time of the short CL, received indomethacin or not. Primary outcome was PTB !35 weeks. Results: Three of the 4 randomized trials identified had databases which recorded indomethacin use at the time of the short CL. A total of 139 women with a short CL !25 mm identified at 14 to 27 weeks were identified. Of these women, 99 (71.2%) received indomethacin, and 40 (28.8%) did not. Demographics and risk factors, including previous PTB (45.5% vs 62.5%; P = .11), were similar in the 2 groups. The primary outcome of PTB !35 weeks occurred in 29.3% (29/99) of women who received indomethacin, and 42.5% (17/40) of women who did not receive indomethacin (RR 0.69, 95% CI 0.44-1.13). PTB !24 weeks occurred in 1.0% (1/99) versus 7.5% (3/40), respectively (RR 0.14; 95% CI 0.02-0.92). Incidence of perinatal death was similar in the 2 groups (6% vs 10%; RR 0.61, 95% CI 0.19-1.95). Conclusion: Indomethacin therapy for asymptomatic women who have a short CL !25 mm on TVU at 14 to 27 weeks and do not receive a cerclage did not prevent spontaneous PTB !35 weeks, but did prevent PTB !24 weeks. Further research including larger numbers and a randomized trial design is necessary to further clarify the effectiveness as well as the risks of this therapy. Ó 2006 Mosby, Inc. All rights reserved.
Preterm birth is the most common cause of perinatal morbidity and mortality. A short cervical length on transvaginal ultrasound is one of the best risk-
assessment methods for prediction of preterm birth.1-3 The appropriate intervention after identifying a woman as high-risk for spontaneous preterm birth because of a
Presented at the Twenty-Sixth Annual Meeting of the Society for Maternal Fetal Medicine, Miami, FL, January 30-February 4, 2006. Reprints not available from the authors. 0002-9378/$ - see front matter Ó 2006 Mosby, Inc. All rights reserved. doi:10.1016/j.ajog.2006.06.007
810 Table I
Berghella, Rust, and Althuisius Descriptive data for each trial
Study
Population screened
Rust6
All singletons, twins, and triplets Singletons with suspected cervical incompetence; twins All singletons Singletons with risk factors for PTB and twins
Althuisius7 To8 Berghella9
GA at screening (wk)
Cervical length (mm)
Funneling (%)
16-24
!25
R 25
14-27
!25
–
22-24 14-24
%15 !25
– O25
PTB, Preterm birth; GA, gestational age; N/A, not available.
sonographic short cervix remains to be determined. Several studies have shown that asymptomatic women with a short cervix on ultrasound are having uterine contractions.4,5 Indomethacin is a nonselective, nonsteroid, anti-inflammatory agent that inhibits the transformation of arachidonic acid into prostaglandins. Prostaglandins are important not only in uterine contractility, but also in the induction of cervical ripening. Cerclage has been so far the only intervention studied to prevent preterm birth once a short cervical length has been detected in asymptomatic women.6-9 Only 1 of these randomized trials found a significant decrease in preterm birth before 34 weeks in favor of the cerclage group.7 In that trial all women treated with a cerclage received indomethacin, while the women not treated with a cerclage did not. This implicates that the indomethacin might have contributed to the positive outcome in the cerclage group. The objective of this study is to systematically review all randomized trials of asymptomatic women with a short cervical length on second-trimester transvaginal ultrasound, examining the effect of indomethacin on prevention of preterm birth.
Material and methods All randomized trials including women with a short cervical length were searched in MEDLINE, PUBMED, EMBASE, and Cochrane Library from 1966 until 2004, inclusive. All manuscripts were reviewed for pertinent references. Selection included asymptomatic pregnant women, found by second-trimester transvaginal ultrasound to have a short cervical length. Exclusion criteria were major fetal anomaly, twin gestation, and cerclage. Only randomized study design trials were included. The eligible trials identified randomized such women to receive either cerclage or no cerclage. Only women who did NOT receive cerclage were analyzed in this study. We compared demographics, risk factors, and outcomes in women who, at the time of the short cervical length, received indomethacin or not. Use of cerclage, tocolytics, or any other intervention around the time of short cervical length was assessed. A metaanalysis on the effect of cerclage on incidence of preterm
birth has already been published.10 Given that characteristics and outcomes reported differed among the manuscripts, primary authors of each trial selected for inclusion were asked for the actual databases, to allow for a patient-level meta-analysis. Primary outcome was PTB !35 weeks. Data analysis was accomplished by relative risks, with 95% confidence intervals. Secondary outcomes reported include incidence of preterm birth at different gestational age cut-offs, incidences of preterm premature rupture of membranes (PPROM), and of perinatal mortality (defined as the sum of intrauterine fetal demise R20 weeks and neonatal demise). Statistical analysis included t test for continuous variables and chi-square or Fisher exact test for categorical variables, and was performed using SPSS, version 12 (Chicago, IL).
Results Four trials6-9 met the inclusion criteria of intervention for short cervical length on second-trimester transvaginal ultrasound (Table I). All primary authors agreed to provide us with the actual databases, including all women randomized. Two authors6,7 supplied data on more randomized women than in their original publications. Three trials recorded use of indomethacin at the time of the short cervical length,6,7,9 and 1 did not.8 The 3 trials included 354 women who had a short cervical length between 14 and 27 weeks. After excluding women who received a ultrasound-indicated cerclage (n = 177), women with multiple gestations (n = 26), and those who received a physical exam-indicated (rescue) cerclage (n = 12), a total of 139 women with singleton gestation, a short CL !25 mm identified at 14 to 27 weeks and no cerclage were identified and analyzed. Of these women, 99 (71.2%) received indomethacin, and 40 (28.8%) did not. All the women in the indomethacin group received indomethacin 100 mg per rectum, followed by 50 mg orally every 6 hours for 48 to 72 hours right after the short CL !25 mm was identified. The majority of women in the indomethacin group came from 1 study,6 and the majority of no indomethacin women from another.9 Demographics and risk factors, including previous PTB, were similar in the
Berghella, Rust, and Althuisius Table I
811
(Continued)
Study
Total sample size (n)
Pericerclage tocolytics (indomethacin)
Rust6
241
Both groups
Althuisius7
52
To8 Berghella9
253 61
Cerclage group only Per obstetrician Per obstetrician
Singleton with no cerclage and indomethacin information (n)
Indomethacin dose Indomethacin 100 mg per rectum, followed by 50 mg orally q6 hours for 48-72 hours Indomethacin not given to no cerclage group N/A Indomethacin 100 mg per rectum, followed by 50 mg orally q6 hours for 48 hours
96 (indomethacin) 14 (no indomethacin) N/A 29 (3 indomethacin/26 no indomethacin)
PTB, Preterm birth; GA, gestational age; N/A, not available.
Table II
Demographics and risk factors
Previous PTB 16-36 wk Cone O1 D&E Short CL (mm)
Table III
Outcomes
Indomethacin (n = 99) (%)
No indomethacin (n = 40) (%)
P value
47 (47)
25 (63)
.11
PTB !35 wk
29 (29)
17 (43)
24 (24) 20 (20) 17.5 G6.6
4 (10) 13 (33) 18.0 G6.8
.06 .12 .70
PTB !32 wk
20 (20)
14 (35)
PTB !24 wk
1 (1)
3 (8)
10 (10)
15 (38)
6 (6)
4 (10)
PTB, Preterm birth; D&E, dilatation and evacuation; CL, cervical length.
2 groups (Table II). The primary outcome of PTB !35 weeks occurred in 29% of women who received indomethacin, and 43% of women who did not receive indomethacin (RR 0.69, 95% CI 0.44-1.13). Incidences of PTB !24 weeks and PPROM were decreased in the indomethacin compared to the control group, while perinatal death was similar in the 2 groups (Table III). In the subgroup of women with previous preterm birth, preterm birth !35 weeks occurred in 34% (16/47) of women who received indomethacin, and 48% (12/25) of women who did not receive indomethacin (RR 0.71, 95% CI 0.42-1.28).
Comment Indomethacin therapy for asymptomatic women who have a short CL !25 mm on TVU at 14 to 27 weeks and do not receive a cerclage was not associated with a significant effect on PTB !35 weeks, but was associated with prevention of both preterm birth !24 weeks and of PPROM. Perinatal mortality was also not significantly different, with a trend toward benefit. For the trend in decrease in preterm birth !35 weeks to reach statistical significance, 200 women with similar characteristics in each group would be needed, assuming an alpha of 0.05 and beta of 0.8. Further research is necessary to clarify the effectiveness as well as the risks of this therapy. There are no other studies on the effect of indomethacin for prevention of preterm birth in asymptomatic women with a short cervical length. In this patient population, the only intervention studied so far has
Indomethacin No indomethacin RR (n = 99) (%) (n = 40) (%) (95% CI)
PPROM Perinatal death
0.69 (0.44-1.13) 0.58 (0.33-1.04) 0.14 (0.02-0.92) 0.27 (0.13-0.54) 0.61 (0.19-1.95)
PTB, Preterm birth; PPROM, premature preterm rupture of membranes; RR, relative risk; CI, confidence interval.
been cervical cerclage. A recent meta-analysis demonstrated prevention of preterm birth in singleton gestations with a short cervical length if cerclage was placed, especially in such women who also had a previous preterm birth.10 In women with a short cervical length but instead carrying twin gestations, cerclage was associated with an increase in preterm birth compared to no cerclage. The results of this meta-analysis should be confirmed by a randomized study. Women who received cerclage for the short cervical length (ultrasound-indicated cerclage) were excluded from our study, to avoid this counfounder. There is insufficient evidence to assess if women who receive cerclage for short cervical length benefit from indomethacin therapy. Almost all women randomized to cerclage in the 4 trials6-9 published so far also received indomethacin, so that no control group (cerclage but no indomethacin) was available for comparison. No studies have been reported on this clinical scenario. Women screened by scheduled transvaginal ultrasound report no symptoms before being identified to have a short cervical length before 24 weeks. The etiology of the short cervical length has been postulated to be either cervical insufficiency,11 infection,12 contractions,4,5 or a combination of these or other factors. Only women with some cervical insufficiency component responsible for the short cervical length might benefit
812 from cerclage. When placed on the tocomonitor after the short cervix has been detected, contractions can easily and frequently be detected in the majority of these asymptomatic women.4,5 This association between short cervix and uterine contractions does not clarify yet if the contractions ‘cause’ the short cervix, or vice versa, or if these changes both derive from a common factor. The fact that contractions are present, though, could make the intervention of tocolysis hypothetically and potentially advantageous. Indomethacin has been studied extensively as a tocolytic in symptomatic women. Compared to placebo, indomethacin is associated with a significant reduction in preterm birth !37 weeks as well as a trend towards a reduction in delivery within 48 hours of initiation of treatment.13 Compared with betamimetics or magnesium, cyclo-oxygenase (COX) inhibitors such as indomethacin are associated with a significant reduction in the number of women delivering within 48 hours and in the incidence of preterm birth !37 weeks.13 In all trials, there was only 1 case of antenatal closure of the ductus arteriosus in 403 women who received short-term tocolysis (up to 48 hours) with COX inhibitors (mainly indomethacin). There was no increase in the incidence of patent ductus arteriosus postnatally (8 treated with COX inhibitors vs 8 treated with placebo or other tocolytics).13 No difference in incidences of intraventricular hemorrhage, bronchopulmonary dysplasia, patent ductus arteriosus, necrotizing enterocolitis, or perinatal mortality was noted in a review aimed at evaluating safety of 1621 fetuses exposed to indomethacin in trials and observational studies compared to controls.14 It would seem reasonable that a short course (48 hours) of indomethacin in the second trimester, before 24 weeks, would not be associated with severe fetal or neonatal complications, but future research should closely assess the safety and risks of this intervention. Strengths of our study are the novel intervention in this population of women at risk for preterm birth because of a short cervical length while asymptomatic. Data analyzed were from prospective, randomized studies, and represent the only assessment of this therapy for this indication thus far in the literature. A shortcoming might have been the sample size, which, if larger, might have lead to different results (eg, significance for preterm birth at !32 weeks). The fact that criteria for administration were either based on trial design6,7 or per obstetrician9 may be another limitation. Another limitation is the fact that long-term infant outcome could not be obtained. The fact that perinatal mortality was nonsignificantly decreased by about 39% in the indomethacin group, with only 1 such outcome in this group, is reassuring. This study shows that indomethacin is associated with a trend toward statistical significance with a 31%
Berghella, Rust, and Althuisius decrease in preterm birth !35 weeks, and with a significant 86% decrease in preterm birth !24 weeks in women with a short cervix before 24 weeks. These results should serve as pilot data for future placebocontrolled trials to assess if indomethacin is beneficial in preventing preterm birth in asymptomatic women with short cervical length on transvaginal ultrasound and either no cerclage or a cerclage in place. Other interventions such as antibiotics, progesterone, bed rest, etc, should also be studied in the future with proper trials. Trials addressing the efficacy of indomethacin should include a specific asymptomatic population (eg, women with previous preterm birth), accurately define short CL (eg, !25 mm), and gestational age (eg, 14-23 6/7 weeks, or even later), dose and duration of therapy (eg, 50 mg every 6 hours for 48 hours), and outcome (eg, preterm birth !35 weeks). Probably this trial is justified both in women with or those without a history-indicated cerclage in place. Other interventions such as those mentioned above should be standardized, and/or controlled for at randomization. Even if indomethacin with or without cerclage is confirmed to benefit certain populations of pregnant women, the incidence of preterm birth with a short cervical length remains high. Other interventions or combinations of interventions should also be tested in the future with properly designed trials.
References 1. Iams JD, Goldenberg RL, Meis PJ, Mercer BM, Moawad A, Das A, et al. The length of the cervix and the risk of spontaneous premature delivery. N Engl J Med 1996;334:567-72. 2. Berghella V, Daly SF, Tolosa JE, DiVito MM, Chalmers R, Garg N, et al. Prediction of preterm delivery with transvaginal ultrasonography of the cervix in patients with high-risk pregnancies: does cerclage prevent prematurity. Am J Obstet Gynecol 1999; 181:809-15. 3. Owen J, Yost N, Berghella V, Thom E, Swain M, Dildy GA, et al. Mid-trimester endovaginal sonography in women at high risk for spontaneous preterm birth. JAMA 2001;286:1340-8. 4. Berghella V, Iams JD, Newman RB, MacPherson C, Goldenberg RL, Mueller-Heubach E, et al. Frequency of uterine contractions in asymptomatic pregnant women with or without a short cervix on transvaginal ultrasound scan. Am J Obstet Gynecol 2004;191: 1253-6. 5. Lewis D, Pelham JJ, Done E, Sawhney H, Talucci M, Berghella V. Uterine contractions in asymptomatic women with a short cervix on ultrasound. J Matern Fetal Neonatal Med 2005;18: 325-8. 6. Rust OA, Atlas RO, Reed J, van Gaalen J, Balducci J. Revisiting the short cervix detected by transvaginal ultrasound in the second trimester: why cerclage therapy may not help. Am J Obstet Gynecol 2001;185:1098-105. 7. Althuisius SM, Dekker GA, Hummel P, Bekedam DJ, van Geijn HP. Final results of the cervical incompetence prevention randomized cerclage trial (CIPRACT): therapeutic cerclage with bed rest versus bed rest alone. Am J Obstet Gynecol 2001;185: 1106-12.
Berghella, Rust, and Althuisius 8. To MS, Alfirevic Z, Heath VCF, Cacho AM, Williamson PR, Nicolaides KH. Cervical cerclage for prevention of preterm delivery in women with short cervix: randomized controlled trial. Lancet 2004;363:1849-53. 9. Berghella V, Odibo AO, Tolosa JE. Cerclage for prevention of preterm birth in women with a short cervix found on transvaginal ultrasound examination: a randomized trial. Am J Obstet Gynecol 2004;191:1311-7. 10. Berghella V, Obido AO, To MS, Rust OA, Althiusius SM. Cerclage for short cervix on ultrasound: meta-analysis of trials using individual patient-level data. Obstet Gynecol 2005;106:181-9. 11. Iams JD, Johnson FF, Sonek J, Sachs L, Gebauer C, Samuels P. Cervical competence as a continuum: a study of ultrasonographic
813 cervical length and obstetric performance. Am J Obstet Gynecol 1995;172:1097-106. 12. Guzman ER, Shen-Schwarz S, Benito C, Vintzileos AM, Lake M, Lai YL. The relationship between placental histology and cervical ultrasongraphy in women at risk for pregnancy loss and spontaneous preterm birth. Am J Obstet Gynecol 1999; 181:793-7. 13. King J, Flenady V, Cole S, Thornton S. Cyclo-oxygenase (COX) inhibitors for treating preterm labour. Cochrane Database of Systematic Reviews 3; 2005. 14. Loe SM, Sanchez-Ramos L, Kaunitz A. Assessing the neonatal safety of indomethacin tocolysis: a systematic review with metaanalysis. Obstet Gynecol 2005;106:173-9.