Objective
The objective of the study was to determine the clinical significance of amniotic fluid (AF) sludge in twin pregnancies with a short cervix.
Study Design
We evaluated twin pregnancies with a short cervical length that had an ultrasound between 16 and 26 weeks (n = 78). Pregnancy outcomes in those with sludge (n = 27) and those without (n = 51) were compared. Outcome variables included gestational age at delivery, premature rupture of the membranes, chorioamnionitis, funisitis, composite neonatal morbidity, and perinatal death. For statistical analysis, the first-born (A) and second-born (B) twins were studied separately.
Results
The prevalence of AF sludge was 34.6% (27 of 78). Pregnancies with sludge delivered earlier (27.2 ± 5.6 weeks vs 31.0 ± 4.05 weeks, P < .01) and had a higher rate of extreme prematurity (<26 weeks: 52.2% [12 of 23] vs 15.6% [5 of 32]; P < .01). Both twins had higher rates of histological chorioamnionitis (twin A, 50.0% [13 of 26] vs 12.8% [6 of 47]; P < .01; twin B, 42.3% [11 of 26] vs 13.3% [6 of 45]; P < .01) and neonatal death (twin A, 33.3% [9 of 27] vs 3.9% [2 of 51]; P < .01; twin B, 33.3% [9 of 27] vs 6.0% [3 of 50]; P = .01). Higher rates of funisitis (23.1% [6 of 26] vs 4.3% [2 of 47]; P = .02) and composite neonatal morbidity were observed for twin A only (66.7% [14 of 21] vs 37.5% [18 of 48]; P = .04).
Conclusion
The presence of AF sludge in twin pregnancies with a short cervix is a risk factor for extreme prematurity, histological chorioamnionitis, and perinatal death. Twin A had higher rates of funisitis and neonatal morbidity in the presence of AF sludge.
Preterm delivery and low birthweight remain the leading causes of neonatal morbidity and mortality in the United States. Multifetal gestations represent a growing group of pregnancies at high risk for preterm delivery. The twin birth rate in the United States increased 76% over the past 30 years, from 18.9 per 1000 in 1980 to 33.2 per 1000 in 2009.
This increase has been largely attributed to the expanded use of fertility-enhancing therapies. Approximately 59% of twins deliver preterm (<37 weeks of gestation) and 11% very preterm (<32 weeks of gestation). In addition, 56.6% of the twins currently born in the United States are considered low birthweight (less than 2500 gams) or very low birthweight (<1500 g).
The higher incidence of prematurity and lower birthweight make twins 8 times more likely to die within the first month of life in addition to having a 4-fold increase in risk of cerebral palsy. Advances in neonatal care have been vast, resulting in a drop in the age of viability to less than 24 weeks and increased survival rates of extremely preterm infants (<28 weeks). Despite this progress, preterm birth remains a public health concern, accounting for 1 in 5 children with mental retardation, 1 in 3 children with visual impairment, and approximately 1 in 2 children with cerebral palsy.
Since the late 1980s, ultrasonographic evaluation of the cervical length (CL) and measurement of fetal fibronectin in cervical secretions have been the main methods used to identify pregnancies at risk for preterm delivery. The association between a second-trimester short CL measured by transvaginal ultrasonography and preterm delivery has been confirmed in all populations studied, including multifetal pregnancies. A recent metaanalysis showed that a CL less than 25 mm measured between 20 and 24 weeks of gestation was strongly associated with preterm delivery (pooled positive likelihood ratio for preterm delivery <28 weeks of 9.6).
Amniotic fluid (AF) sludge has been proposed as an additional ultrasound marker for preterm delivery. Espinoza et al coined this term in 2005 to describe dense aggregates of particulate matter that can be identified in close proximity to the internal cervical os and have imaging characteristics similar to gallbladder sludge seen by ultrasound.
AF sludge has been studied in women with preterm labor with intact membranes, asymptomatic women at high risk for preterm delivery, and asymptomatic women undergoing amniocentesis for fetal karyotyping. It has consistently been found to represent an independent risk factor for early preterm delivery, premature rupture of the membranes (PROM), chorioamnionitis, funisitis, neonatal morbidity, and neonatal mortality in these populations.
The effect of sludge on twin pregnancy outcomes is unknown. The objective of this study was to evaluate the risk of preterm delivery in twin pregnancies with and without AF sludge. We also sought to investigate the relationship between the presence of sludge and the incidence of chorioamnionitis, funisitis, and adverse neonatal outcomes in twin pregnancies.
Materials and Methods
A retrospective cohort study was conducted of twin pregnancies referred to the Fetal Imaging Unit of Beaumont Health System between January 2004 and February 2011. The study was approved by the Human Investigation Committee of Beaumont Health System.
Pregnancies that had an ultrasonographic cervical length measurement 25 mm or less and at least 1 of the ultrasonographic measurements performed between 16 and 26 weeks were included (n = 88). None of the patients had uterine contractions and/or cervical dilatation by digital examination at the time of the examination. Ten pregnancies with 1 or more of the following conditions were excluded: prenatal diagnosis of chromosomal and/or morphological anomalies (n = 3), intrauterine fetal demise (n = 3), and loss to follow-up (n = 4), leaving 78 pregnancies for final analysis.
The presence of AF sludge was recorded in the ultrasound database but not used for clinical care. Transvaginal ultrasound was conducted using commercially available ultrasound systems (Acuson Sequoia; Siemens Healthcare, Mountain View, CA; Voluson 730; GE Healthcare, Milwaukee, WI; and Philips IU-22; Philips Healthcare, Bothell, WA). Endovaginal transducers with a frequency ranging from 4 to 9 MHz were used. Registered diagnostic medical sonographers performed the ultrasound examinations and all transvaginal cervical examinations were performed by sonologists (either board-certified perinatologists or radiologists).
Cervical length was measured from the external os to the internal cervical os. AF sludge was defined as the presence of dense aggregates of particulate matter in proximity to the internal cervical os as originally described by Espinoza et al. One of the authors (L.G.), who participated in previous studies of AF sludge and who was blinded to the clinical outcomes, reviewed all static images and video clips stored in a picture archiving and communication system (KinetDx; Siemens Healthcare) or videotapes to confirm the presence or absence of AF sludge for all cases. The presence of sludge was tabulated only as present or not present between 16 and 26 weeks, regardless of when the shortest cervical length was measured.
Gestational age was determined by the last menstrual period or by the in vitro fertilization transfer date and confirmed by a first-trimester ultrasound. The hospital electronic medical records were queried for maternal characteristics, labor and delivery records, and pregnancy outcomes. Pregnancy outcomes in those with (n = 27) vs. without (n = 51) AF sludge were compared. Gestational age at the diagnosis of a short CL was defined as the earliest gestational age at which a cervical length of 25 mm or less was documented.
Primary outcome variables included preterm delivery at less than 26 weeks, 26-32 weeks, and greater than 32 weeks’ gestation. Gestational age at delivery was analyzed only when the delivery was spontaneous (n = 55) and not indicated by maternal or fetal illness.
Secondary outcomes included PROM, clinical and histological chorioamnionitis, funisitis, admission to the neonatal intensive care unit (NICU), composite neonatal morbidity, and perinatal death. Composite neonatal morbidity was defined as the presence of any of the following: respiratory distress syndrome, bronchopulmonary dysplasia, intraventricular hemorrhage, patent ductus arteriosus, necrotizing enterocolitis, and hyaline membrane disease.
For statistical analysis, each twin was studied individually (n = 78 for each analysis), and then the outcomes for the first-born (A) and second-born (B) twins were compared.
Comparisons among groups were performed using a χ 2 or Fisher exact tests for categorical variables and a Wilcoxon rank-sum test for continuous variables. Comparisons between twins were performed using the McNemar’s test to account for the correlation between siblings. All reported P values are 2 sided. Odds ratios (ORs) and corresponding 95% confidence intervals (CIs) were calculated for variables that showed an initial statistical significance according to a value of P < .05. An exact Kruskal-Wallis test was performed to evaluate the distribution of gestational age at delivery according to the presence or absence of AF sludge.
Logistic regression analysis was performed to adjust for confounding factors such as cervical length and shortest cervical length observed. All analyses were performed using SAS version 9.3 (SAS Institute Inc, Cary, NC), StatXact version 9 (Cytel Inc, Cambridge, MA), and STATA version 12 (StataCorp LP, College Station, TX).
Results
The prevalence of AF sludge in twin pregnancies was 34.6% (27 of 78). The demographic characteristics of the study population according to the presence or absence of AF sludge are shown in Table 1 . There were no differences in maternal age, parity, prepregnancy body mass index, racial distribution, rates of smoking, uterine anomalies or previous loop electrosurgical excision procedure surgery, prior history of dilation and curettage, prior second-trimester loss, prior preterm delivery, prior cesarean section, or conception through in vitro fertilization.
| Demographic | No sludge, (n = 51) | Sludge, (n = 27) | P value |
|---|---|---|---|
| Maternal age, y | 32.3 ± 6.0 | 32.1 ± 4.3 | .86 |
| Gravidity | |||
| 1 | 16/51 (31.4%) | 6/27 (22.2%) | |
| 2 | 18/51 (35.3%) | 13/27 (48.2%) | |
| 3 | 3/51 (19.6%) | 4/27 (14.8%) | |
| >3 | 7/51 (13.7%) | 4/27 (14.8%) | |
| Parity | |||
| 0 | 37/51 (72.6%) | 18/27 (66.7%) | |
| 1 | 13/51 (25.5%) | 5/27 (18.5%) | |
| 2 | 1/51 (1.9%) | 3/27 (11.1%) | |
| 3 | — | 1/27 (3.7%) | |
| Race | |||
| White | 16/20 (80%) | 12/17 (70.6%) | |
| African-American | 2/20 (10%) | 2/17 (11.8%) | |
| Hispanic | 1/20 (5%) | 0 | |
| Asian | 1/20 (5%) | 3/17 (17.7%) | |
| BMI (prepregnancy) | 26.8 ± 4.5 | 29.7 ± 7.8 | .11 |
| Smoking | 2/27 (7.4%) | 3/14 (21.4%) | |
| Uterine anomalies and/or LEEP | 3/50 (6.0%) | 2/27 (7.4%) | 1.00 |
| History of D&C | 11/40 (27.5%) | 5/25 (20.0%) | .57 |
| History of second-trimester loss | 2/41 (4.9%) | 4/25 (16.0%) | .19 |
| History of preterm delivery | 5/47 (10.6%) | 5/26 (19.2%) | .31 |
| Prior cesarean section | 4/51 (7.8%) | 2/27 (7.4%) | 1.00 |
| IVF | 22/39 (56.4%) | 11/21 (52.4%) | .79 |
Pregnancy characteristics are shown in Table 2 . Most pregnancies had dichorionic-diamniotic (84.4%), followed by monochorionic-diamniotic (14.3%), and monochorionic-monoamniotic (1.3%) placentation. There was no difference in the incidence of twin-to-twin transfusion syndrome, maternal bleeding during pregnancy, or cerclage placement.
| Characteristics | No sludge, (n = 51) | Sludge, (n = 27) | P value |
|---|---|---|---|
| Placentation | .83 | ||
| Dichorionic-diamniotic | 42/51 (82.4%) | 23/26 (88.5%) | |
| Monochorionic-diamniotic | 8/51 (15.7%) | 3/26 (11.5%) | |
| Monochorionic-monoamniotic | 1/51 (2.0%) | 0 | |
| Twin-to-twin transfusion | 2/51 (3.9%) | 1/27 (3.7%) | 1.00 |
| Bleeding | 12/50 (24.0%) | 9/27 (33.3%) | .43 |
| Cerclage | 3/51 (5.9%) | 3/27 (11.1%) | .41 |
| Shortest CL between 16 and 26 wks, cm a | 2.2 (1.5–3.1) | 0.6 (0.0–1.2) | < .01 |
| GA at CL of ≤15 mm, wks a | 25.1 (24.1–27.3) | 21.2 (19.6–23.3) | < .01 |
| GA at CL of ≤10 mm, wks a | 25.1 (24.1–25.0) | 21.3 (20.0–23.3) | < .01 |
a Continuous variables are reported as median (interquartile range) and tested by the Wilcoxon rank-sum test.
The gestational age at which a short cervix of 15 mm or less or of 10 mm or less was first seen was significantly lower among patients with sludge. The shortest cervical length measured between 16 and 26 weeks was also shorter in the group with sludge (0.6 cm; interquartile range, 0.0–1.2 cm vs 2.2 cm; interquartile range, 1.5–3.1 cm; P < .01). The proportion of pregnancies with a cervical length of 10 mm or less was significantly higher among patients with AF sludge ( Figure ).
Pregnancies with AF sludge delivered significantly earlier (median, 25.7 weeks; interquartile range, 22.6–32.7 weeks vs median, 31.9 weeks; interquartile range, 28.6–33.9 weeks; P < .01). The rate of preterm delivery rate for each gestational age group below 30 weeks was significantly higher for the sludge group ( Table 3 ). More than half of the deliveries in the sludge group occurred before 26 weeks (52.2% [12 of 23] vs 15.6% [4 of 32]; P = .002), whereas in pregnancies without sludge, a 50% delivery rate was reached only at 32 weeks. The rate of PROM was similar in pregnancies with and without AF sludge.
| Clinical characteristics | No sludge, (n = 51) | Sludge, (n = 27) | P value | Odds ratio (95% CI) |
|---|---|---|---|---|
| PROM | 20/51 (39.2%) | 11/27 (40.7%) | 1.00 | — |
| Gestational age at spontaneous delivery, wks a,b | 31.9 (28.6–33.9) | 25.7 (22.6–32.7) | .01 | — |
| Delivery <32 weeks a | 16/32 (50.0%) | 17/23 (73.9%) | .10 | 2.83 (0.78–10.97) |
| Delivery <30 wks a | 11/32 (34.4%) | 17/23 (73.9%) | .006 | 5.41 (1.45–21.3) |
| Delivery <28 wks a | 6/32 (18.8%) | 14/23 (60.9%) | .002 | 6.75 (1.72–27.63) |
| Delivery <26 wks a | 4/32 (15.6%) | 12/23 (52.2%) | .02 | 7.63 (1.745–38.19) |
a Analysis was restricted to pregnancies that delivered spontaneously (no indicated deliveries such as fetal distress, severe maternal disease, placental abruptio, etc were included). No sludge group included 32 subjects, and the sludge group included 23 subjects. Data were missing for 1 patient in the no sludge group
A multivariate linear regression model was fitted to the data to explore the relationship between CL, AF sludge, and gestational age at delivery. Variables included in the model were selected using stepwise regression. The presence of AF sludge, the gestational age at which the shortest CL was observed, the shortest CL observed, and the interaction between the absence of sludge and shortest CL observed were all statistically significant ( Table 4 ). Cervical lengths of 25 mm or less, 15 mm or less, and 10 mm or less were analyzed, and a CL of 10 mm or less interacted the most strongly with AF sludge.
| Parameter | Estimate | SE | t value | P value | 95% CI |
|---|---|---|---|---|---|
| Intercept | 15.66 | 4.67 | 3.36 | .0016 | 6.27–25.05 |
| AF sludge, no | –8.75 | 2.837 | -3.09 | .0034 | –14.45 to –3.05 |
| AF sludge, yes | 0.00 | — | — | — | — |
| Minimum CL age | 1.00 | 0.18 | 5.43 | < .0001 | 0.63–1.37 |
| Shortest CL obs CL 10 | –11.23 | 2.65 | –4.24 | .0001 | –16.57 to –5.90 |
| Shortest CL obs CL 15 | –8.93 | 3.57 | –2.50 | .0159 | –16.11 to –1.75 |
| Shortest CL obs CL 25 | 0.00 | — | — | — | — |
| AF sludge shortest CL obs no CL 10 | 9.35 | 3.06 | 3.05 | .0038 | 3.18–15.51 |
| AF sludge shortest CL obs no CL 15 | 6.52 | 4.00 | 1.63 | .1099 | –1.53 to 14.57 |
| AF sludge shortest CL obs no CL 25 | 0.00 | — | — | — | — |
| AF sludge shortest CL obs yes CL 10 | 0.00 | — | — | — | — |
| AF sludge shortest CL obs yes CL 15 | 0.00 | — | — | — | — |
| AF sludge shortest CL obs yes CL 25 | 0.00 | — | — | — | — |
Differences in the rates of intraamniotic inflammation were evaluated by comparing the rates of histological chorioamnionitis and funisitis for both twins ( Table 5 ). The rate of histological chorioamnionitis was higher when sludge was present (twin A: 50.0% [13 of 26] vs 12.8% [6 of 47]; P = .0008; twin B: 42.3% [11 of 26] vs 13.3% [6 of 45]; P = .007). A higher rate of chorioamnionitis for twin A was seen, regardless of cervical length ( Table 6 ), with an OR of 4.76 (95% CI, 1.06–24.25). Logistic regression using the variables AF sludge, CL of 10 mm or less and shortest CL observed demonstrated that AF sludge is an independent predictor for histological chorioamnionitis for twin A (sludge: coefficient estimate, 1.60, SE, 0.69, and P = .021; CL, ≤10 mm: coefficient estimate, –0.031, SE 0.72, and P = .669).
| Clinical characteristics | Twin | No sludge present, (n = 51) | Sludge present, (n = 27) | P value | OR (95% CI) | McNemar’s |
|---|---|---|---|---|---|---|
| Birthweight, g | A | 1854 ± 716 | 1209 ± 851 | .002 | — | |
| B | 1722 ± 745 | 1213 ± 794 | .008 | |||
| Apgar 1 min a | A | 8 (8–9) | 6 (3–8) | .001 | — | |
| B | 8 (6-8) | 6 (3–8) | .043 | |||
| Apgar 5 min a | A | 9 (9–9) | 8 (7–9) | < .001 | — | |
| B | 9 (8–9) | 8 (7–9) | .16 | |||
| NICU days | A | 17 (6–40.5) | 48.5 (4–98) | .19 | — | |
| B | 18 (7–38) | 57 (7–101) | .045 | |||
| Discolored amniotic fluid | A | 3/51 (5.9%) | 2/27 (7.4%) | 1.00 | 0.20 | |
| B | 3/44 (6.8%) | 0/21 (0%) | .55 | |||
| Clinical chorioamnionitis | A | 4/47 (8.5%) | 5/26 (35.6%) | .27 | 1.00 | |
| B | 4/45 (8.9%) | 5/26 (9.2%) | .27 | |||
| Histological chorioamnionitis | A | 6/47 (12.8%) | 13/26 (50.0%) | .001 | 0.69 | |
| B | 6/45 (13.3%) | 11/26 (42.3%) | .006 | |||
| Funisitis | A | 2/47 (4.3%) | 6/26 (23.1%) | .021 | 6.75 (1.06–71.92) | 0.22 |
| B | 2/46 (4.4%) | 2/26 (7.7%) | .62 | 1.79 (0.12–25.96) | ||
| NICU admission | A | 38/49 (77.6%) | 21/26 (80.1%) | 1.00 | 1.00 | |
| B | 38/45 (84.4%) | 7/24 (70.8%) | .22 | |||
| Sepsis | A | 4/48 (8.3%) | 1/20 (5%) | 1.00 | 0.63 | |
| B | 3/46 (6.5%) | 4/21 (19.1%) | 1.00 | |||
| RDS | A | 5/48 (10.4%) | 2/21 (9.5%) | 1.00 | 0.22 | |
| B | 7/46 (15.2%) | 3/21 (14.3%) | 1.00 | |||
| HMD | A | 14/48 (29.2%) | 14/21 (66.7%) | .007 | 4.86 (1.43–17.16) | 0.27 |
| B | 21/47 (44.7%) | 13/22 (59.1% | .31 | 1.79 (0.57–5.72) | ||
| BPD | A | 3/48 (6.3%) | 4/21 (19.1%) | .19 | 0.56 | |
| B | 3/46 (6.5%) | 2/20 (10.0%) | .63 | |||
| IVH | A | 2/48 (4.2%) | 1/21 (4.8%) | 1.00 | 0.73 | |
| B | 2/46 (4.4%) | 3/21 (14.3%) | .32 | |||
| PDA | A | 9/48 (18.8%) | 8/21 (38.1%) | .13 | 1.00 | |
| B | 10/46 (21.7%) | 8/21 (31.3%) | .23 | |||
| NEC | A | 0/48 (0%) | 2/48 (9.5%) | .09 | 1.00 | |
| B | 1/46 (2.2%) | 1/20 (5%) | .52 | |||
| Composite neonatal morbidity b | A | 18/48 (37.5%) | 14//21 (66.7%) | .036 | 3.33 (1.009–11.552) | 0.021 b |
| B | 25/47 (53.2%) | 16/22 (72.7%) | .19 | 2.35 (0.70–8.56) | ||
| Neonatal death | A | 2/51 (3.9%) | 9/27 (33.3%) | < .001 | 12.25 (2.16–122.50) | 1.00 |
| B | 3/50 (6.0%) | 9/27 (35.1%) | .003 | 7.83 (1.65–45.54) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
