Introduction
The obstetrical patient presenting to labor and delivery triage for a labor evaluation at term is one of the most common clinical scenarios. Currently, the diagnosis of true labor at term relies on patient symptomatology and very frequently on progressive cervical dilation by digital vaginal exams. However, perception of contractions is a poor predictor of labor and digital exams and therefore Bishop score have large intraobserver and interobserver variabilities, providing low accuracy to predict true labor.
There are several studies examining the use of cervical length (CL) surveillance by transvaginal ultrasound (TVUS) to predict spontaneous preterm birth in symptomatic as well as asymptomatic patients. As a matter of fact, the usefulness of CL to predict preterm labor (PTL) has been documented very well so that it is now routinely used in clinical practice. In term patients the use of CL has been limited to prediction of spontaneous labor in prolonged pregnancies and also in the prediction of successful labor induction.
To our knowledge, there are no data on CL measurements to differentiate true from false labor in term patients presenting for a labor check. Consequences from false-positive diagnosis of true labor at term are unnecessary hospital admissions, unnecessary obstetrical interventions, increased resource utilization, and increased cost. Therefore, the primary objective of this prospective study was to determine if CL by TVUS can differentiate true from false labor in term patients who present to the hospital for labor check. Our secondary objective was to determine the relationship between CL and time to delivery in these patients.
Materials and Methods
This was an institutional review board–approved prospective observational study from 2013 through 2016 in term (37-42 weeks) patients presenting to labor and delivery triage at Winthrop University Hospital, Mineola, NY, with labor symptoms. A history and physical was performed as standard of care. Informed consent was obtained after determining eligibility. Recruitment was not consecutive secondary to time constraints because of a very busy labor and delivery unit where the study was conducted. Inclusion criteria were: singleton, live intrauterine pregnancy in cephalic presentation, gestational age 37-42 weeks, regular uterine contractions (defined as ≥4 contractions/20 min on the tocometer), intact membranes, and cervix <4 cm dilated and <80% effaced. Exclusion criteria were: clinical chorioamnionitis (defined as temperature >100.4°F and 2 of the following: malodorous discharge, maternal leukocytosis, maternal tachycardia, fetal tachycardia, uterine tenderness), maternal or fetal indications for immediate delivery, placenta previa, and previous cesarean delivery.
TVUS was performed by residents previously trained on proper CL technique. Three images per patient were collected and the shortest best image was chosen to be analyzed in the study. Providers making management decisions were blinded to the CL measurements. True labor was defined as spontaneous rupture of membranes or spontaneous cervical dilation ≥4 cm and ≥80% effaced within 24 hours of CL measurement. False labor was defined as cases that did not fulfill the above definition of true labor. Patient’s demographics and obstetrical variables were collected and included: patient’s age, prepregnancy body mass index, race, gravidity, parity, gestational age, cervical exam, CL measurement upon presentation, date and time of CL measurement, date of admission to or discharge from hospital, time of active labor, induction or augmentation of labor, spontaneous or artificial rupture of membranes and timing, time of birth, birthweight, and mode of delivery.
Several statistical analyses were performed. The demographic and clinical characteristics of true vs false labor patients were compared with parametric and nonparametric analysis. Receiver operating characteristic (ROC) curves of CL in the prediction of true labor were generated and the diagnostic accuracy of CL was determined for various CL cutoffs for both primiparous and multiparous patients. The diagnostic accuracy of the various CL cutoffs was expressed by sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and positive and negative likelihood ratios for both primiparous and multiparous patients separately as well as combined. Lastly, the relationship between CL and time to delivery was determined by linear regression and scatter plot and the correlation coefficient (r) was calculated. CL vs time to delivery was analyzed individually for patients who received oxytocin augmentation and those who did not.
Materials and Methods
This was an institutional review board–approved prospective observational study from 2013 through 2016 in term (37-42 weeks) patients presenting to labor and delivery triage at Winthrop University Hospital, Mineola, NY, with labor symptoms. A history and physical was performed as standard of care. Informed consent was obtained after determining eligibility. Recruitment was not consecutive secondary to time constraints because of a very busy labor and delivery unit where the study was conducted. Inclusion criteria were: singleton, live intrauterine pregnancy in cephalic presentation, gestational age 37-42 weeks, regular uterine contractions (defined as ≥4 contractions/20 min on the tocometer), intact membranes, and cervix <4 cm dilated and <80% effaced. Exclusion criteria were: clinical chorioamnionitis (defined as temperature >100.4°F and 2 of the following: malodorous discharge, maternal leukocytosis, maternal tachycardia, fetal tachycardia, uterine tenderness), maternal or fetal indications for immediate delivery, placenta previa, and previous cesarean delivery.
TVUS was performed by residents previously trained on proper CL technique. Three images per patient were collected and the shortest best image was chosen to be analyzed in the study. Providers making management decisions were blinded to the CL measurements. True labor was defined as spontaneous rupture of membranes or spontaneous cervical dilation ≥4 cm and ≥80% effaced within 24 hours of CL measurement. False labor was defined as cases that did not fulfill the above definition of true labor. Patient’s demographics and obstetrical variables were collected and included: patient’s age, prepregnancy body mass index, race, gravidity, parity, gestational age, cervical exam, CL measurement upon presentation, date and time of CL measurement, date of admission to or discharge from hospital, time of active labor, induction or augmentation of labor, spontaneous or artificial rupture of membranes and timing, time of birth, birthweight, and mode of delivery.
Several statistical analyses were performed. The demographic and clinical characteristics of true vs false labor patients were compared with parametric and nonparametric analysis. Receiver operating characteristic (ROC) curves of CL in the prediction of true labor were generated and the diagnostic accuracy of CL was determined for various CL cutoffs for both primiparous and multiparous patients. The diagnostic accuracy of the various CL cutoffs was expressed by sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and positive and negative likelihood ratios for both primiparous and multiparous patients separately as well as combined. Lastly, the relationship between CL and time to delivery was determined by linear regression and scatter plot and the correlation coefficient (r) was calculated. CL vs time to delivery was analyzed individually for patients who received oxytocin augmentation and those who did not.
Results
In all, 101 patients were enrolled; 24 patients were excluded because of: induction prior to active labor (22), breech presentation after the CL measurement (delivered by scheduled cesarean) (1), and spontaneous rupture of membranes after the vaginal exam and before the TVUS (1). Analysis was performed on the remaining 77 patients.
Of the 77 patients analyzed, 45 were in true labor (prevalence of true labor 45/77 or 58.4%) and 32 (41.6%) were in false labor. The groups were similar in regards to maternal age, body mass index, race, gravidity, parity, and mode of delivery ( Table 1 ). Patients who were in true labor had shorter CL measurements at the time of presentation as compared to those in false labor: median 1.3 cm (range 0.5-4.1) vs 2.4 cm (range 1.0-5.0), respectively ( P < .001). Those who were in true labor were also of more advanced gestational age compared to the false labor patients ( P < .01) ( Table 1 ).
| True labor n = 45 [58.4%] | False labor n = 32 [41.6%] | P value | |
|---|---|---|---|
| Age, y | 28.7 ± 6.0 | 27.7 ± 5.7 | .47 |
| BMI, kg/m 2 | 29.9 ± 5.5 | 30.1 ± 5.0 | .85 |
| Race | .71 | ||
| White | 20 (44.4) | 12 (37.5) | |
| Black | 13 (28.9) | 12 (37.5) | |
| Hispanic | 12 (26.7) | 5 (15.6) | |
| Other | 0 (0.0) | 3 (9.4) | |
| Gravidity | 2 (1–5) | 2 (1–6) | .54 |
| Parity | 0 (0–3) | 0 (0–2) | .30 |
| Gestational age, wk | 39.3 ± 0.9 | 38.7 ± 1.1 | <.01 |
| Cesarean delivery | 7 (15.6) | 5 (15.6) | 1.00 |
| Birthweight, g | 3285 ± 434 | 3413 ± 375 | .18 |
| Cervical length, cm | 1.3 (0.5–4.1) | 2.4 (1.0–5.0) | <.001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
