Recent reports reveal a disturbing increase in pregnancy-related mortality in the United States (CDC Website, 2016). Similarly, the rate at which infants are being born far too early is equally shocking, and has been getting worse, not better (CDC Website, 2018) in the last two years. In fact, maternal mortality and premature delivery rates are both higher in the US than in any other high-income country and are actually similar to those in some developing countries (CDC Website, 2017; March of Dimes, 2018).

With over 700 women dying of pregnancy-related causes each year (CDC Website, 2016) and more than 1,000 babies born prematurely EACH DAY (March of Dimes, 2016), there is a pressing need to reassess and update our approach to the care for pregnant women. One area offering an opportunity for improvement is the assessment of pregnant women who present with signs and symptoms of preterm labor (PTL).

 

The Clinical Impact of Failing to Accurately Identify Women at Risk of PTL

When clinicians fail to fully evaluate symptoms suggesting PTL, they miss an opportunity to act on behalf of both mothers and their unborn babies. Substantial research supports the need to transfer expectant mothers at high risk of spontaneous preterm birth (sPTB) to tertiary care centers, where they can receive antenatal corticosteroids, magnesium sulfate and antibiotics to optimize neonatal outcomes when preterm delivery occurs despite preventive measures.

Failure to accurately triage pregnant women with symptoms of PTL increases the likelihood that those at high risk for sPTB will deliver prematurely in a facility that is unprepared to care for critically-ill infants. Such failures increase the probability of short- and long-term complications for the child, which substantially raise healthcare expenditures.

Approximately 4 million babies are born annually in the US, with around 10% of them being born preterm (<37 weeks of gestation) (CDC Website, 2018). Of those preterm births, approximately 30% are medically-indicated or due to identifiable maternal diagnoses such as preeclampsia or to fetal complications such as intrauterine fetal growth restriction. The remaining 70% are spontaneous preterm births caused by premature rupture of membranes or PTL (Goldenberg et al, 2008).

Given the raw numbers, identifying women with PTL symptoms who will ultimately deliver preterm is essential. However, accurate assessment of PTL risk remains a challenge for healthcare providers. Clinical criteria for diagnosis – physical exam findings such as cervical dilation and regular uterine contractions – are unreliable, as 90% of women considered likely to deliver preterm based on these criteria will not deliver in 7 days of their exam, and 75% will deliver at term (DeFranco et al, 2013; Fuchs et al, 2004).

 

The Need for Standardized Protocols to Evaluate Women With Signs of PTL

Since classic clinical criteria for the diagnosis of PTL are non-specific and poorly predictive of sPTB, there is a clear need for protocols endorsing routine use of tools with both strong positive and negative predictive values, such as fetal fibronectin (fFN) testing and transvaginal ultrasound measurement of cervical length (TVUS).  However, while professional organizations such as the American College of Obstetricians and Gynecologists (ACOG) have called for the development of standardized protocols (CPSQI, 2015), they have not yet developed or endorsed guidelines specifically related to the evaluation of pregnant women with symptoms of PTL.  Such guidance is direly needed evidenced by a recent study by Blackwell which clearly demonstrates the sizeable inadequacies of obstetric triage units and the potential negative outcomes associated with preterm delivery in the United States.

In the study by Blackwell, a total population of greater than 23,000 pregnant women with symptoms of preterm labor were evaluated using an insurance claims database.  Of the total cohort, 17,512 pregnant women were discharged from the hospital after an evaluation for symptoms of PTL.  Surprisingly, 20% of these pregnant women who were discharged home gave birth within 72 hours of their hospital evaluation.  Of the 3,517 patients who delivered preterm infants within 3 days, fFN testing was used in only 4% of their evaluations.  The use of TVUS in their evaluation was also remarkably low at 18%.  Had clinicians utilized fFN testing as part of a strict protocol for preterm labor triage in this study cohort, undoubtedly many of these high-risk expectant women would have been identified with a positive fFN offering an opportunity for hospitalization and proper treatment with antenatal corticosteroids and magnesium sulfate, both of which have been shown to improve neonatal outcomes in preterm infants.  Additionally, these newborns would have been more likely to be born in a facility well prepared to address potential complications associated with preterm birth (Blackwell et al, 2017).

The desired algorithm(s) to evaluate women with symptoms of PTL should embrace clinical steps already reported in the literature (van Baaren 2014), but with the recognition that evaluation of preterm delivery risk based on symptom assessment and physical examination alone is inaccurate.  Importantly, an algorithm should also note that If an institution has the capability to perform both transvaginal ultrasound and fFN testing, then a strategy should be adhered to for use of both of these diagnostic tools, as the two tests work well together to help reliably distinguish between patients who should be admitted and those who should be discharged.

 

An Algorithm for Triaging Women With Symptoms of PTL

A simple algorithm such as the one proposed here (Figure) can improve the care of women with symptoms of PTL. In this algorithm, Step 1 includes a comprehensive medical examination using clinical criteria set forth in ACOG’s Practice Bulletin no. 171. During Step 2, a speculum examination should be performed, during which swabs for Group B streptococcus, gonorrhea and chlamydia, wet mount, and fFN are taken. Cervical length should be measured using transvaginal ultrasound (TVUS) in facilities staffed to do so and cervical dilation should then be measured by digital examination.

During Step 3, in women with cervical dilation <3 cm, the decision to admit or discharge should be based on cervical length (if TVUS was performed) combined with fFN results. Expectant mothers with cervical length <15 mm or 15-30 mm and a positive fFN test should be admitted or transferred to an appropriate facility for monitoring and treatment. Patients with cervical length >30 mm or 15-30 mm plus a negative fFN test may be discharged. If TVUS is not performed, women with cervical dilation <3 cm on digital examination should immediately undergo fFN testing. Patients with negative fFN findings have a <1% chance of delivering within the next 14 days and can be discharged with confidence (fFN PI, 2017).

Standardized Algorithms for PTL Triage Improve Pregnancy Outcomes

A similar algorithm implemented at the Mayo Clinic called for the use of fFN testing for 201 women with a cervical length between 16-29 mm and cervical dilation ≤2 cm. The algorithm showed a negative predictive value for delivery within 7 and 14 days from evaluation of 99.2% and 97.2%, respectively, and reduced maternal admissions by 56% while substantially lowering costs (Rose et al, 2010).

Another algorithm adopted at Brigham & Women’s Hospital in Boston helped reduce annual hospital days for unnecessary maternal admissions from unindicated maternal admissions from 1853 to 903 days after implementation of fFN testing alone. A savings of nearly $1,000,000 was realized in unnecessary medical expenses for the hospital (Iyer et al, 2013).

Given the implications for both maternal and neonatal outcomes, fFN testing should be routinely incorporated with TVUS to evaluate all expectant mothers with signs of PTL, as the two tests together provide the greatest predictive ability for preterm birth. Clinicians should be encouraged to adhere to standardized algorithms for triage in at-risk mothers in order to maximize the utility of available diagnostic tools and reduce the impact of PTL and preterm births both on patients and the healthcare system.

References

American College of Obstetricians and Gynecologists – Obstetrics. Practice Bulletin no. 171: management of preterm labor. ACOG Practice Bulletin No. 171. Obstet Gynecol. 2016;128:e155-e164.

Blackwell SC, Sullivan ER, Petrilla AA, Troeger KA, Byrne JD. Utilization of fetal fibronectin testing and pregnancy outcomes among women with symptoms of preterm labor. Clinicoecon Outcomes Res. 2017; 9:585-594.

CDC. Maternal Health. https://www.cdc.gov/chronicdisease/resources/publications/aag/maternal.htm. Updated October 27, 2017. Accessed November 20, 2018.

CDC. Meeting the Challenges of Measuring and Preventing Maternal Mortality in the United States. https://www.cdc.gov/grand-rounds/pp/2017/20171114-maternal-mortality.html. Updated February 28, 2018. Accessed November 20, 2018.

CDC. Premature birth. www.cdc.gov/features/prematurebirth/index.html. Updated November 5, 2018. Accessed November 20, 2018

Committee on Patient Safety and Quality Improvement. Committee Opinion No. 629: clinical guidelines and standardization of practice to improve outcomes. Obstet Gynecol. 2015;125(4):1027-1029.

Iyer S, McElrath T, Jarolim P, Greenberg J. The association of fFN testing on hospital admissions for preterm labor. Open J Obstet Gynecol. 2013; 3:126-129.

Rose CH, McWeeney DT, Brost BC, et al. Cost-effective standardization of preterm labor evaluation. Am J Obstet Gynecol. 2010;203:250.e1-5.

van Baaren GJ, Vis, JY, Wilms, FF, et al. Predictive value of cervical length measurement and fibronectin testing in threatened preterm labor. Obstet Gynecol. 2014;123(6):1185-92.