Materials and Methods
All patients examined at Rochester Methodist Hospital Obstetrics Triage between December 2007 and November 2008 (inclusive) underwent evaluation according to a standardized protocol ( Figure 1 ) as validated by Schmitz et al and Hedriana and Bliss, who were endorsed by both the March of Dimes and the Society for Maternal-Fetal Medicine. Briefly, after gestational age was confirmed and contraction frequency was established, a speculum examination was performed to obtain routine data on gonorrhea, chlamydia, and group B streptococcal cultures and fetal fibronectin. This was followed by a digital cervical examination and cervix length measurement with transvaginal ultrasound scanning; subsequent obstetrics decisions regarding management, corticosteroid administration, and hospitalization for tocolysis were based on these parameters. If the cervix length measurement was ≥3cm, cultures and fetal fibronectin were discarded; if there was not clinical concern for abruptio placentae or chorioamnionitis, the patient was discharged. All patients who were transported from an outlying facility were admitted for a 23-hour observation period. Obstetric demographic characteristics, incidences of deviation from protocol, hospital admissions, gestational age at both evaluation and delivery, and interval elapsed time between evaluation and delivery were recorded for all patients. This study was approved by the Mayo Clinic Institutional Review Board under protocol #10-000887.
After an initial 30-day introductory period, the protocol was implemented as the exclusive method for the triage of patients with symptoms of preterm labor. All patients who were transferred to our institution from outlying facilities underwent a similar evaluation; although the women would be admitted for 23-hour observation, for clinical purposes they were treated identically. All clinical examinations were performed by both house staff and faculty.
All patient triage visits (approximately 4500 per year) were reviewed prospectively by the primary author (C.H.R.), and pregnancy outcome data were retrieved from the electronic medical record where available. Protocol violations were recorded, and the responsible physicians were contacted directly for verbal case review. Final appropriateness of hospital admission was determined after review of admission criteria, discussion with attending staff, and review of hospital course.
The historical comparison group was comprised of patients who had been admitted from January 1 to December 31, 2006, who were treated with conventional methods for the exclusion of preterm labor (ie, continuous electronic monitoring with serial cervical examinations). Decisions regarding hospitalization were individualized and left to the discretion of individual attending physicians. Numeric admission data were obtained for comparative purposes, and costs of hospitalization were based on the estimated duration of hospitalization of 48 hours at a charge of $2100 per day.
Results
During the initial 12-month study interval, 201 patients had 215 visits to Obstetrics Triage for suspected preterm labor. Protocol violations occurred in 48 instances (22%) because fetal fibronectin was sent before cervix length measurement (40%), because no cervical length measurement was available (27%), because neither cervix length nor fetal fibronectin was performed (27%), or because of other reasons (6%; Figure 2 ). Outcome data were unavailable for 21 patients (10%), primarily because of transport status and subsequent delivery at their respective referring institutions. Mean gestational age at delivery was 38 weeks 3 days, with an average interval of 57.4 days between evaluation and delivery ( Table 1 ). Fifteen patients were admitted over the study interval, compared with 34 patients during preceding year; however, 11 of these 15 patients met criteria for admission according to the protocol. Only 3 patients (1.5%) delivered within 7 days of evaluation ( Figure 3 ), with the following characteristics:
| Variable | Measure |
|---|---|
| Patients included, n | 201 |
| Protocol violations, n | 48 |
| Outcome data unavailable, n | 21 |
| Patients admitted, n | 15 |
| Delivery within 7 days of evaluation, n | 3 |
| Delivery within 14 days of evaluation, n | 6 |
| Mean gestational age at delivery, wk | 38 4/7 |
| Mean duration between evaluation and delivery, d | 57.4 |
