in 1992. Numbers of all fellowship programs and positions and allapplicants and positions filled were recorded annually. We plottedapplicant and position data by year to examine trends over time.RESULTS: The number of applicants has tripled since 1996 with moreapplicants and applicants per position than ever before. The numbersof participating programs, positions offered, and applicants declinedduring the transition period between 1997 and 2003. The number ofparticipating fellowship programs rose steadily thereafter and is nowequivalent to numbers before instituting the additional training year.While the number of positions increased in recent years, it has notreached levels before 1996. In the last nine years, the competition foracceptance into a fellowship in MFM has risen; there has been a meanof 42.2 � 17.1 unmatched applicants per year. Comparatively, from1992-1996 and during the expansion period (1997-2003) the meannumber of unmatched applicants was 14.7 � 4.5 and 12.3 � 5.9 re-spectively. Similarly, from 2004-2012 there has been a mean of 3.8 �3.0 unfilled programs per year versus 28.7 � 2.6 from 1992-1996 and23.7 � 10.7 from 1997-2003.CONCLUSION: The transition from a two to a three year training pro-gram in 1996 only temporarily halted the otherwise steady and sus-tained growth in MFM fellowship programs, positions, and appli-cants.

540 Training needs in operative obstetrics amongfirst year maternal-fetal medicine fellowsKacey Eichelberger1, Sue Tolleson-Rinehart3, Angie Bentson2,Kate Menard1

1University of North Carolina, Chapel Hill, Obstetrics & Gynecology, ChapelHill, NC, 2University of North Carolina, Chapel Hill, Epidemiology, ChapelHill, NC, 3University of North Carolina, Chapel Hill, Public Health, ChapelHill, NCOBJECTIVE: Toestimatethefrequencywithwhich13operativeobstetricpro-cedures were performed by first year MFM fellows during their precedingresidencies,andtodeterminefactorsassociatedwithincreasedcomfortdoingand teaching these procedures during fellowship.STUDY DESIGN: We used Qualtrics software to administer a web-basedsurvey to all 100 first-year fellows, and exact logistic regression toestimate the odds of comfort in doing and teaching these procedures.We present descriptive data on all 13 procedures and additional detailon three common ones: cesarean hysterectomy, low forceps delivery,and breech extraction of a second twin.RESULTS: Response rate was 86%; frequency data for all 13 proceduresis shown in Table 1. Fellows who completed residency in the North-east/Midatlantic (N�26) were less likely to report comfort doing orteaching low forceps deliveries (OR 0.21 [0.05,0.78], and 0.20[0.04,0.85], respectively), while those completing fellowship in theWest (N�13) reported more comfort performing breech extractionof a second twin (OR 6.84 [1.24, 51.50]); fellows completing residencyin the Southeast formed the referent group (N�26). Fellows reportingcompletion of the three selected procedures � 5 times each duringresidency were significantly more likely to report comfort doing and

teaching them as fellows. Type of residency program (community/academic) was not significantly associated with reported comfort, al-though size of program was: fellows from residency programs with 5-9residents/ year (N�57) were less likely to report comfort teachingbreech extraction of a second twin (OR 0.18 [0.03, 0.76]) or low for-ceps delivery (0.27 [0.07, 1.02]) than were those from programs of lessthan 5 residents per year.CONCLUSION: The wide range of operative obstetric experience fel-lows reported gaining in residency varies by program demo-graphic. To attain goals outlined in ABOG’s Guide to Learningrelated to operative obstetrics, many MFM fellows need additionalprocedural experience.

541 Estimating blood loss: can an easily accessible visualaid significantly improve visual estimation?Lisa Zuckerwise1, Cheryl Raab1, Catalin Buhimschi1, ChristianPettker1, Heather Lipkind1

1Yale University, Obstetrics, Gynecology & Reproductive Sciences, NewHaven, CTOBJECTIVE: Obstetric hemorrhage is an emergency situation that re-quires expedient identification and intervention to prevent maternaland fetal morbidity. Research has shown that visual estimation ofblood loss (EBL) is inaccurate and often underestimated, which maylead to delayed treatment. Our objective was to determine whether aneasily accessible visual aid would improve provider’s performance inthe estimation of blood loss.STUDY DESIGN: We created a pocket card, depicting known volumes ofblood on common obstetric materials, to serve as a visual aid in esti-mating blood loss (Figure). Accuracy of blood loss estimation wasassessed at 6 stations with known volumes of artificial blood usingmaterials common to standard delivery kits and recreating commonobstetric scenarios. Medical personnel consisting of 151 subjects [ob-stetrics attending (n�24), residents (n�22), nurses (n�74), mid-wives (n�11), anesthesia providers (n�13), medical students (n�7)]recorded EBL across a variety of volumes and materials before andafter receiving our visual aid. Clinical experience of the evaluatorsranged from 0-40 years (Mean 11.54 � SD 11.89). We prospectivelycategorized percent error of EBL as underestimation (% error lessthan �20%), accurate assessment (% error between �20% and�20%); and overestimation (% error more than �20%). Wilcoxonsigned rank test was used for analysis.RESULTS: For all providers, there was a significant improvement in theaccuracy of EBL after intervention across all blood volumes testedexcept for the Chux A (Table). In other analyses, initial accuracy ofEBL was significantly affected by provider type in only 2 of the 6stations (Peri-pad A and Delivery Drape) (p � 0.05). This difference

Operative obstetric frequency data for firstyear Maternal-Fetal Medicine fellows

Column 1 includes data from 86 respondents; columns 2-4 include data from 85 respondents.

Poster Session IV Academic Issues, Antepartum Fetal, Clinical Ob, Fetus, Genetics, Hypertension, Med-Surg-Diseases, Operative Ob, U/S www.AJOG.org

S232 American Journal of Obstetrics & Gynecology Supplement to JANUARY 2013

persisted in only one of the stations after the intervention (DeliveryDrape) (p � 0.01).CONCLUSION: A visual aid depicting known volumes of blood on com-mon obstetric materials leads to improved accuracy of EBL and mayserve as a rapid decision-making tool for obstetric providers.

542 Availability of dilation and evacuation trainingat maternal-fetal medicine fellowshipsMelissa Rosenstein1, Jody Steinauer1, Aaron Caughey2, JenniferKerns1

1University of California, San Francisco, Obstetrics, Gynecology, andReproductive Sciences, San Francisco, CA, 2Oregon Health and ScienceUniversity, Obstetrics & Gynecology, Portland, OROBJECTIVE: To describe the availability of second-trimester dilationand evacuation (D&E) training opportunities at US maternal-fetalmedicine (MFM) fellowships.STUDY DESIGN: This is a cross-sectional survey of all of all ABOG-approved MFM fellowships in the United States. We contacted pro-gram directors via email and invited them to complete a 10 questionweb-based survey. Names and email addresses were found on thepublicly-available SMFM Fellowship Directory. Questions addressedthe availability and uptake of D&E training in fellowship, as well asgeographic and institutional characteristics of the training environ-ment. Participants were offered a $5 gift card as remuneration.

RESULTS: Of the 76 program directors contacted, 40 (53%) responded tothe survey. The 40 programs represented have a total of 158 fellows en-rolled (median 3 fellows per program, range 1-13). The fellowships rep-resented by the survey respondents have a similar geographic distributionto the distribution of all US MFM fellowships; 38% of respondents arelocated in the Northeast, 30% in the South/Southwest, 18% in the Mid-west, and 15% in the West. The availability of D&E training varies greatly;7 programs (18%) offer routine training, 12 programs (30%) offer orga-nized optional training, and16 programs (40%) do not have formalizedtraining opportunities but allow fellows to participate in D&Es that areoccasionally performed. Five programs (13%) have no D&E trainingavailable. Fellowship directors at programs without formalized routine oroptional training opportunities are more likely to report that a majority oftheir fellows pursue D&E training (8/21 vs 4/19, p�0.07).CONCLUSION: SomeformofD&Etraining isavailableatmostof theMFMfellowship sites participating in this survey. Program directors report thatmany fellows are motivated to acquire skills in D&E, and expanding thesetraining opportunities could be beneficial to MFM fellows.

543 Immune-modulatory effect of chronic resveratroltreatment on LPS induced inflammation inmurine pregnancyRamzy Nakad1, George Saade1, Maged Costantine1, MonicaLongo1, Talar Kechichian1, Michel Makhlouf1

1UTMB Galveston, Maternal fetal medicine, Galveston, TXOBJECTIVE: Resveratrol (RV) is a polyphenol secreted by plants in responseto stress; it had been recognized for its anti-senesce properties. Several mech-anismshadbeensuggestedforthisanti-agingeffectsuchasSIRT-1activation,antioxidant,anti-inflammatorypropertiesandimprovingvascularfunction.Anti-inflammatory properties such as decreasing COX-2 and i-NOS pro-duction; blocking TNFa trigger NFkb activation preventing IL6, COX2 pro-duction.RVdecreasesexpressionofTNFa,IL6,IL1b,ICAM-1andiNOS.RVprevents LPS induced lethality in mice. The main purpose of this study is toevaluatetheeffectofResveratrolonpreventingLPSinducedinflammationinthe uterine and fetal brain unit.STUDY DESIGN: Pregnant CD1 mice, were injected daily with RV8mg/kg (n�8) vs placebo (n�8) intraperitoneally starting at Day#7till Day #15.On Day #15 the mice were randomized to receive 0.1 ml ofLPS vs control.On Day #16 Dams were euthanized with carbon diox-ide. Uterine muscle tissue and fetal brains were obtained. Uterinemuscle tissue was processed and western blots were prepared to mea-sure Connexin43, TNFa and COX2. Fetal brains were harvested PCRwas used to measure IL1B, TNFa and IL6 levels. The results wereevaluated using Student t test and analysis of Variance for multiplecomparisons. P � 0.05 was considered statistically significantRESULTS: There was no significant difference in the maternal weightgain, fetal weights, placental weights or number of pups between theRV and control group. Treatment with RV had no significant anti-inflammatory effect in the LPS model.CONCLUSION: Chronic intraperitoneal resveratrol treatment seems notto have an anti-inflammatory effect at the given dose.

Visual aid with measured volumes of blood

Participants categorized by percent errorof EBL before and after intervention

* Underestimation (% error less than �20%), accurate assessment (% error between �20% and�20%), and overestimation (% error more than �20%); †Wilcoxon Signed Rank Test.

All values are N (%) unless otherwise specified.

www.AJOG.org Academic Issues, Antepartum Fetal, Clinical Ob, Fetus, Genetics, Hypertension, Med-Surg-Diseases, Operative Ob, U/S Poster Session IV

Supplement to JANUARY 2013 American Journal of Obstetrics & Gynecology S233