INTRODUCTION Globally, preterm birth is signicant maternal and neonatal problems that is a leading cause of neonatal death and disability. The incidence of premature birth is estimated about 15 million annually and has increased in all countries .(Sharvit et al., 2017) This condition causes many problems to the survival neonates such as vision, respiratory, neurological and developmental problems (Nirunsittirat et al., 2016; Society for Maternal-Fetal Medicine . Electronic address, McIntosh, Feltovich, Berghella, & Manuck, 2016). The diagnosis and management of this condition are still problematic such as the diagnostic criteria, tocolytics use. Cervical length (CL) is an indicator to diagnose preterm labor in developed countries (Sotiriadis, Papatheodorou, Kavvadias, & Makrydimas, 2010; Tsoi, Akmal, Rane, Otigbah, & Nicolaides, 2003; Tsoi, Fuchs, Rane, Geerts, & Nicolaides, 2005). The American College of Obstetricians and Gynecologists (ACOG) recommended the use of CL for the diagnosis and treatment of preterm labor. The short CL < 3 cm is an indication for further investigation by perform fetal bronectin and CL < 2cm is an indication for preterm labor treatment (American College of & Gynecologists' Committee on Practice, 2016). However, many centers in low-resource countries cannot provide CL measurement from many reasons such as lack of trained personnel or lack of equipment. In many hospitals, ultrasound machine is available without vaginal probe. Therefore, this study was conducted to determine the diagnostic performance in using abdominal convex ultrasonic probe for TPS-CL compare with TVS-CL, in case of TVS-CL is not available. MATERIAL AND METHOD This diagnostic test study was conducted in a tertiary care regional hospital from December 2018 through May 2019. This study protocol was approved by the Ethical Research Committee of Udonthani Hospital (No.40/2561). The inclusion criteria were: 1) women with singleton live preterm (GA 25-36 weeks) pregnancies, 2) who had painful and regular uterine contraction of at least one time in every 10 minutes persisting for more than 30 minutes (Chawanpaiboon, Pimol, & Sirisomboon, 2011) 3). The exclusion criteria were cervix dilated more than 3 cm, placenta previa, abnormal vaginal bleeding, preterm premature rupture of membranes, dead fetus in utero, fetal anomaly, uncertain gestational age, history of previous cervical or uterine surgery, the clear image of TPS-CL or TVS-CL cannot be obtained or unwilling to participate in the study. The eligible participants were counseled and invited to participate in this study. A written informed consent was obtained after the explanation of study methods to the participants. Then, all participants were examined by digital vaginal examination, by experienced nurses who have worked in the labor room for at least 5 years, to rule out advance labor (cervical dilatation > 3 cm). First, CL was measured transperineally, then transvaginally by trained obstetricians or residents, using a GE Voluson P6 ultrasound machine with a 3.5-5 MHz convex probe and 7-10 MHz transvaginal probe. The patient was examined with an empty bladder. The CL was obtained by measuring from external to internal os in a longitudinal axis. Three measurements were taken and the shortest values was taken as the nal CL in both measurements. Transperineal measurement was done by apply a convex ultrasonic probe, which was covered by a sterile plastic bag, at anterior perineum and between labia majora (Raungron gmorakot et al., 2004). The ultrasonic probe was moved until clear image found. Cut off value for short CL diagnosis was < 30 mm (American College of & Gynecologists' Committee on Practice, 2016; Ness, Visintine, Ricci, & Berghella, 2007). All DIAGNOSTIC PERFORMANCE OF TRANSPERINEAL SONOGRAPHIC CERVICAL LENGTH MEASUREMENT FOR DIAGNOSIS OF PRETERM LABOR: A DIAGNOSTIC TEST STUDY Original Research Paper Jirada Promnimit M.d. Department of Obstetrics and Gynecology, Udonthani Hospital, Udonthani, Thailand. Obstetrics & Gynecology OBJECTIVE: To evaluate the diagnostic performance of transperineal sonographic cervical length measurement (TPS-CL) compare with transvaginal measurement (TVS-CL). MATERIALS AND METHODS: The participants were 107 suspected preterm labor women whom cervical dilatation was < 3 cm. TPS-CL and TVS-CL were measured. The short cervical length (CL) was diagnosed when CL < 3 cm. The diagnostic performance of TPS-CL was evaluated. RESULTS: Short CL (CL < 3cm) was found in 52.3% by TVS-CL and 51.4% by TPS-CL. TPS-CL had high correlation with TVS-CL (r = 0.77). The sensitivity, specicity, positive and negative predictive value, and accuracy of TPS-CL were 78.6% (95%CI 65.6- 88.4), 78.4% (95%CI 64.7-88.7), 80.0% (95%CI 67.0- 89.6), 76.9%% (95%CI 63.2-87.5), 78.5 % (95%CI 69.5-85.9), respectively. CONCLUSION: TPS-CL had high correlation with TVS-CL. In case of TVS-CL is unavailable, TPS-CL can be used for preterm labor diagnosis. ABSTRACT KEYWORDS : Cervical Length, Suspected Preterm Labor, Transperineal Ultrasound, Transvaginal Ultrasound Srisuda Songthamwat M.d. Department of Obstetrics and Gynecology, Udonthani Hospital, Udonthani, Thailand. Ueamporn Summart Ph.d. Faculty Of Health Sciences, Udon Thani Rajabhat University, Udonthani, Thailand. Metha Songthamwat M.d., Ph.d. Department of Obstetrics and Gynecology, Udonthani Hospital, Udonthani, Thailand. *Corresponding Author VOLUME-8, ISSUE-12, DECEMBER-2019 • PRINT ISSN No. 2277 - 8160 • DOI : 10.36106/gjra 60 X GJRA - GLOBAL JOURNAL FOR RESEARCH ANALYSIS staff and residents, who participated in this study, were trained and assessed for standardization of CL measurement in a 1-day course. Then, the participants were managed according to the hospital protocol and the Royal Thai College of Obstetricians and Gynecologists (RTCOG) guideline (The Royal Thai College of the Obstetricians and Gynaecologists, 2015). STATISTICAL ANALYSIS The sample size was calculated using the formula for diagnostic test study. The estimated sensitivity of TPS-CL was 80 % with 0.08 acceptable error and 5% signicance were used. The calculated sample size was 97, then a 10% dropout rate was added. The total sample size was 107 (Khiewyoo, 2014; Sanoonrat, Srisantiroj, & Yanaranop, 2017). This study protocol was approved by the Research Ethics Committee of Udonthani Hospital (number No.40/2561). The written informed consent was obtained from all participants before enrolling to this study. The participants' characteristics are presented in term of frequency and percentage for categorical data. Mean, standard deviation and range were used for continuous data. TVS-CL < 2 cm was used as the gold standard diagnosis of short CL. Diagnostic performance of TPS-CL was assessed including sensitivity, specicity, positive and negative predictive value and accuracy. Statistical analysis was performed using Stata program version13. P-value <0.05 was considered statistically signicant. RESULTS There were 116 participants with suspected threatened preterm labor were examined by TPS-CL and TVS-CL and 9 participants were excluded due to unable to obtain the clear image of TPS-CL. The clear image of TPS-CL and TVS-CL can be measured in 107 (92.2%) and 100.0% of participants, respectively. Their mean age was 25.6 (±SD 7.0) years and 47 (43.9%) women were primigravida. Mean gestational age was 33.2 (±SD 2.5) weeks. Baseline characteristics are presented in Table1. Table 1. Basic characteristic of study group. The mean TVS-CL was 3.0 cm (±SD 0.92) and mean TPS-CL was 3.0 cm (+SD 0.94). The Figure 1 show the correlation of TPS-CL and TVS-CL. Signicant correlation of TPS-CL and TVS-CL was demonstrated ( r = 0.77) Abbreviation: CL ; cervical length Figure 1: The correlation of transperineal and transvaginal cervical length measurement The prevalence of short CL (TVS-CL < 3 cm) was 52.3% (95%CI: 42.5-62.1), range of CL was 0.4-5.3 cm. TPS-CL found short CL in 51.4%. The sensitivity, specicity, positive and negative predictive value and accuracy were 78.6% (95%CI 65.6-88.4), 78.4% (95%CI 64.7-88.7), 80.0% (95%CI 67.0- 89.6), 76.9%% (95%CI 63.2-87.5), 3.6 (95%CI 2.1-6.3), 0.27 (95%CI 0.16-0.46), 78.5 % (95%CI 69.5-85.9), respectively. (Table 2). Table 2.: Diagnostic performance of transperineal sonog raphic measurement for short cervical length (< 3 cm) compare with standard transvaginal measurement. Abbreviation: TVS: transvaginal sonography, TPS: transper ineal sonography, CL: cervical length, CI: condence interval DISCUSSION Preterm labor is sometime difcult to differentiate from false labor pain or Braxton-hicks contraction, especially in case of minimal cervical dilatation. CL measurement has been recommended by many obstetric guidelines as a part of preterm labor diagnosis and management in developed countries (American College of & Gynecologists' Committee on Practice, 2016; Lockwood C, 2018; National Institute for Health and Care Excellence, 2018). However, in low and middle-income countries, an ultrasound machine with transvaginal probe and trained personnel are still unavailable in many hospitals. In some hospitals, ultrasound with convex probe is available but transvaginal probe is not. Data from this study demonstrated that CL measurement by TPS-CL had a high correlation with TVS-CL and can be used with high sensitivity and specicity when compared with TVS- CL. Therefore, TPS-CL can be used in the situation that VOLUME-8, ISSUE-12, DECEMBER-2019 • PRINT ISSN No. 2277 - 8160 • DOI : 10.36106/gjra Characteristics Total (N=107) (%) Maternal age (years), n(%) Mean ±SD < 20 20 – 34 ≥ 35 25.6+7.0 26 69 12 24.3% 64.5% 11.2% Gestational age at admission (weeks), n(%) Mean ±SD < 34 weeks >34 weeks 33.2 + 2.5 46 61 43.0% 57.0% Primigravida, n (%) 47 43.9% Body mass index (kg/m2) Mean(±SD) < 18.5 18.5 – 22.9 23 – 24.9 25 – 29.9 >30 25.4 (+4.1) 4 27 21 40 15 3.7% 25.2% 19.6% 37.4% 14.0% Previous preterm birth, n(%) 8 7.5% Previous term birth, n(%) 42 39.3% Initial dilatation of cervix (cm) Closed Dilate ≥ 1 cm 63 44 58.9% 41.1% Bishop score Mean (±SD) < 6 ≥ 6 3.8 (+2.5) 80 27 74.8% 25.2% TVS-CL TPS_CL Short CL (< 3cm) CL > 3 cm Short CL(< 3cm) 44 11 CL > 3 cm 12 40 Diagnostic performance Percent 95%CI Prevalence of short CL 52.3 42.5-62.1 Sensitivity 78.6 65.6-88.4 Specicity 78.4 64.7-88.7 Positive predictive value 80.0 67.0- 89.6 Negative predictive value 76.9 63.2-87.5 Positive likelihood ratio 3.6 2.1-6.3 Negative likelihood ratio 0.27 0.16-0.46 Accuracy 78.5 69.5-85.9 X 61GJRA - GLOBAL JOURNAL FOR RESEARCH ANALYSIS vaginal ultrasonic probe is unavailable for evaluating the cervical condition of suspected preterm labor patients. Data from this study is compatible with Raungrongmorakot K, et al study (Raungrongmorakot et al., 2004) which found signicant correlation between TPS-CL and TVS-CL ( r = 0.73), Meijer-Hoogeveen M, et al study which found strong correlation with r = 0.85, Larscheid P ,et al study which found high correlation with Lin'rho =0.922. However, this is the rst study that reports the diagnostic performance of TPS-CL in the diagnosis of short CL for preterm labor management. In this study, the denition of short CL was < 3 cm in accordance with ACOG guideline and Ness A, et al study (American College of & Gynecologists' Committee on Practice, 2016; Ness et al., 2007). However, different cut-off lengths of CL were used in other recommendations. National Collaborating Centre for Women's and Children's Health (NICE) guideline (Sarri , Davies, Gholi tabar, Norman, & Guideline Development, 2015), Tsoi E, et al (Tsoi et al., 2003) and Alrevic Z, et al (Alrevic, Allen-Coward, Molina, Vinuesa, & Nicolaides, 2007) study recommends that if CL is > 1.5 cm, the diagnosis of preterm birth is unlikely. Palacio M, et al study (Palacio et al., 2007) reported cut-off length of 25 mm in gestational age less than 32 weeks and 15 mm in 32 weeks or later. Therefore, the diagnostic performance of TPS-CL will be different if others cut-off lengths are used. There are some limitations with this study; rst the standardization of observer was done by a workshop, however the inter and intra-observer reliability of TPS-CL and TVS-CL were not measured, therefore reliability problem can be occurred. Second, the TPS-CL and TVS-CL were done by same operator which can have a measurement bias. The further study should be done to compare the outcome of preterm management between using TPS-CL and TVS-CL in the suspected preterm labor cases. CONCLUSION: TPS-CL had high correlation with TVS-CL and high sensitivity and specicity when compared with TVS-CL. In case of TVS- CL is unavailable, TPS-CL can be used for diagnosis of short CL in preterm labor diagnosis. ACKNOWLEDGEMENTS We gratefully acknowledge Dr.Narong Thadadech, Director of Udonthani Hospital for permission and grant support. Thanks for Udonthani Hospital staff and all participants who participated in this trial. DISCLOSURE STATEMENT No author has any potential conict of interest. FUNDING This study was supported by Udonthani Hospital. The study protocol was approved by the Udonthani Research Ethics Committee: No.40/2561 REFERENCES 1. Alrevic, Z., Allen-Coward, H., Molina, F., Vinuesa, C. P., & Nicolaides, K. (2007). Targeted therapy for threatened preterm labor based on sonographic measurement of the cervical length: a randomized controlled trial. Ultrasound Obstet Gynecol, 29(1), 47-50. doi: 10.1002/uog.3908 2. American College of, O., & Gynecologists' Committee on Practice, B.-O. (2016). Practice Bulletin No. 171: Management of Preterm Labor. Obstet Gynecol, 128(4), e155-164. doi: 10.1097/AOG.0000000000001711 3. Chawanpaiboon, S., Pimol, K., & Sirisomboon, R. (2011). Comparison of success rate of nifedipine, progesterone, and bed rest for inhibiting uterine contraction in threatened preterm labor. Journal of Obstetrics and Gynaecology Research, 37(7), 787-791. 4. Khiewyoo, J. (2014). Statistics for Diagnostic Teast. In J. Khiewyoo (Ed.), Statistical Mrthods for Health Measurement (1 ed., pp. 57-77). Khon Kaen. 5. Lockwood C. (2018). Diagnosis of preterm labor. In: Uptodate. Retrieved May 29, 2018, from https://www.uptodate.com/contents/diagnosis-of-preterm- labor 6. National Institute for Health and Care Excellence. (2018). Preterm labour and birth. Retrieved May 29, 2018, from https://www.nice.org.uk/guidance/ng25 /chapter/Recommendations#diagnosing-preterm-labour-for-women-with- intact-membranes 7. Ness, A., Visintine, J., Ricci, E., & Berghella, V. (2007). Does knowledge of cervical length and fetal bronectin affect management of women with threatened preterm labor? A randomized trial. Am J Obstet Gynecol, 197(4), 426 e421-427. doi: 10.1016/j.ajog.2007.07.017 8. Nirunsittirat, A., Pitiphat, W., McKinney, C. M., DeRouen, T. A., Chansamak, N., Angwaravong, O., . . . Pimpak, T. (2016). Adverse birth outcomes and childhood caries: a cohort study. Community Dent Oral Epidemiol, 44(3), 239- 247. doi: 10.1111/cdoe.12211 9. Palacio, M., Sanin-Blair, J., Sanchez, M., Crispi, F., Gomez, O., Carreras, E., . . . Gratacos, E. (2007). The use of a variable cut-off value of cervical length in women admitted for preterm labor before and after 32 weeks. Ultrasound Obstet Gynecol, 29(4), 421-426. doi: 10.1002/uog.3950 10. Raungrongmorakot, K., Tanmoun, N., Ruangvutilert, P., Boriboonhirunsarn, D., Tontisirin, P., & Butsansee, W. (2004). Correlation of uterine cervical length measurement from transabdominal, transperineal and transvaginal ultrasonography. J Med Assoc Thai, 87(3), 326-332. 11. Sanoonrat, P., Srisantiroj, N., & Yanaranop, M. (2017). Spot Urine Albumin to Creatinine Ratio versus Urine Protein to Creatinine Ratio for the Diagnosis of Proteinuria in Pregnancy. Thai J Obstet Gynaecol, 25(4), 249-258. 12. Sarri, G., Davies, M., Gholitabar, M., Norman, J. E., & Guideline Development, G. (2015). Preterm labour: summary of NICE guidance. BMJ, 351, h6283. doi: 10.1136/bmj.h6283 13. Sharvit, M., Weiss, R., Ganor Paz, Y., Tzadikevitch Geffen, K., Danielli Miller, N., & Biron-Shental, T. (2017). Vaginal examination vs. cervical length - which is superior in predicting preterm birth? J Perinat Med, 45(8), 977-983. doi: 10.1515/jpm-2016-0273 14. Society for Maternal-Fetal Medicine . Electronic address, p. s. o., McIntosh, J., Feltovich, H., Berghella, V., & Manuck, T. (2016). The role of routine cervical length screening in selected high- and low-risk women for preterm birth prevention. Am J Obstet Gynecol, 215(3), B2-7. doi: 10.1016/j.ajog.2016.04.027 15. Sotiriadis, A., Papatheodorou, S., Kavvadias, A., & Makrydimas, G. (2010). Transvaginal cervical length measurement for prediction of preterm birth in women with threatened preterm labor: a meta-analysis. Ultrasound Obstet Gynecol, 35(1), 54-64. doi: 10.1002/uog.7457 16. The Royal Thai College of the Obstetricians and Gynaecologists. (2015). Clinical practice guideline OB014: The Management of Preterm labour and Premature Rupture of Membranes (pp. 15-42). 17. Tsoi, E., Akmal, S., Rane, S., Otigbah, C., & Nicolaides, K. H. (2003). Ultrasound assessment of cervical length in threatened preterm labor. Ultrasound Obstet Gynecol, 21(6), 552-555. doi: 10.1002/uog.131 18. Tsoi, E., Fuchs, I. B., Rane, S., Geerts, L., & Nicolaides, K. H. (2005). Sonographic measurement of cervical length in threatened preterm labor in singleton pregnancies with intact membranes. Ultrasound Obstet Gynecol, 25(4), 353-356. doi: 10.1002/uog.1809 VOLUME-8, ISSUE-12, DECEMBER-2019 • PRINT ISSN No. 2277 - 8160 • DOI : 10.36106/gjra 62 X GJRA - GLOBAL JOURNAL FOR RESEARCH ANALYSIS