Intraobserver and Interobserver Reliability of Transvaginal Cervical Length Measurements and Quantitative Ultrasound Tissue Characterization of the Cervix in the Second and Third Trimester of Pregnancy

Zusammenfassung

Ziel: Bestimmung der Intra- und Interobserver-Reliabilität der transvaginalen Zervixlängenmessung und der quantitativen sonografischen Gewebetypisierung (QUTC). Material und Methoden: Bei 20 Frauen im zweiten und dritten Trimester wurde eine transvaginale Sonografie der Zervix von 2 Untersuchern mit jeweils 3 Untersuchungen pro Patientin durchgeführt. Zervixlänge und QUTC wurden bestimmt. Wiederholbarkeitskoeffizient, Intraclass-Korrelationskoeffizient und Interclass-Korrelationskoeffizient wurden berechnet sowie die Bland-Altman-Methode angewandt. Ergebnisse: Der Wiederholbarkeitskoeffizient für die transvaginale Zervixlängenmessung betrug 0,95 bei Untersucher 1 und 0,99 bei Untersucher 2. Die entsprechenden Intraclass-Korrelationskoeffizienten waren 0,872 und 0,973. Der Interclass-Korrelationskoeffizient für die Zervixlänge betrug 0,986 und die mittlere Interobserver-Differenz war –2 mm. Die Übereinstimmungsgrenzen nach der Bland-Altmann-Methode betrugen –5,4 mm bis 1,5 mm. Der Wiederholbarkeitskoeffizient für die QUTC betrug 0,90 bei Untersucher 1 und 0,97 bei Untersucher 2. Die entsprechenden Intraclass-Korrelationskoeffizienten waren 0,74 und 0,9. Der Interclass-Korrelationskoeffizient für QUTC betrug 0,56 und die mittlere Interobserver-Differenz war 0,14. Die Übereinstimmungsgrenzen nach der Bland-Altman-Methode waren weit und betrugen –22,46 bis 22,74. Schlussfolgerung: Die transvaginale Zervixlängenmessung zeigt eine hohe Intra- und Interobserver-Reliabilität. QUTC zeigt eine nur eine geringe Interobserver-Reliabilität und sollte daher gegenwärtig nicht im klinischen Alltag angewendet werden.

Abstract

Purpose: To determine the intraobserver and interobserver reliability of transvaginal ultrasound measurements of cervical length and quantitative ultrasonic tissue characterization (QUTC). Materials and Methods: 20 healthy women in the second and third trimester of pregnancy underwent transvaginal ultrasound examination of the cervix by two sonographers. 3 replicate vaginal measurements of cervical length and QUTC were taken by each observer for each woman. The repeatability coefficient and intraclass correlation coefficient as well as the interclass correlation coefficient and limits of agreement were calculated. Results: The reliability coefficient for transvaginal cervical length measurements was 0.95 for sonographer 1 and 0.99 for sonographer 2. The corresponding intraclass correlation coefficients for cervical length measurements were 0.872 and 0.973. The interclass correlation coefficient for transvaginal cervical length measurements was 0.986, the mean interobserver difference was –2 mm, and the limits of agreement were –5.4 mm to 1.5 mm. The reliability coefficients for QUTC were 0.90 (sonographer 1) and 0.97 (sonographer 2) and the corresponding intraclass correlation coefficients for cervical length measurements were 0.74 and 0.9. The interclass correlation coefficient was 0.56, the mean interobserver difference was 0.14, and the limits of agreement were –22.46 to 22.74. Conclusion: Transvaginal cervical length measurements showed a high intraobserver and interobserver reliability, while quantitative ultrasonic tissue characterization had poor interobserver reliability and should currently not be applied in clinical practice.

Key words

quantitative ultrasonic tissue characterization - cervical length - intraobserver reliability - interobserver reliability

References

• 1 Iams J. Prevention of preterm birth.  N Engl J Med. 1998;  338 54-56
• 2 Moster D, Lie R L, Markestad T. Long-Term Medical and Social Consequences of Preterm Birth.  N Engl J Med. 2008;  359 262-273
• 3 Kobaly K, Schluchter M, Minich N et al. Outcomes of extremely low birth weight (< 1 kg) and extremely low gestational age (< 28 weeks) infants with bronchopulmonary dysplasia: effects of practice changes in 2000 to 2003.  Pediatrics. 2008;  121 73-81
• 4 Hack M, Taylor H G, Drotar D et al. Chronic conditions, functional limitations, and special health care needs of school-aged children born with extremely low-birth-weight in the 1990 s.  JAMA. 2005;  294 318-325
• 5 Berkowitz G S. An epidemiologic study of preterm delivery.  Am J Epidemiol. 1981;  113 81-92
• 6 Levkoff A H, Westphal M, Miller 3 rd M C et al. Maternal risk factors in infants with very low birth weight.  Obstet Gynecol. 1982;  60 612-616
• 7 Papiernik E, Kaminski M. Multifactorial study of the risk of prematurity at 32 weeks of gestation. I. A study of the frequency of 30 predictive characteristics.  J Perinat Med. 1974;  2 30-36
• 8 Shiono P H, Klebanoff M A, Graubard B I et al. Birth weight among women of different ethnic groups.  JAMA. 1986;  255 48-52
• 9 Iams J D, Goldenberg R L, Meis P J et al. The length of the cervix and the risk of spontaneous premature delivery. National Institute of Child Health and Human Development Maternal Fetal Medicine Unit Network.  N Engl J Med. 1996;  334 567-572
• 10 Kagan K O, To M, Tsoi E et al. Preterm birth: the value of sonographic measurement of cervical length.  BJOG. 2006;  113 52-56
• 11 Jörn H, Kalf K, Schwann R et al. Grey-scale texture analysis of the cervix in pregnancy.  Ultraschall in Med. 2006;  27 347-54
• 12 Heller F, Voigt F, Schrauder M et al. B-Bildanalyse der Zervix vor der medikamentösen Geburtseinleitung – Können Rückschlüsse auf den weiteren Geburtsverlauf gezogen werden?.  Ultraschall in Med. 2010;  31 S01
• 13 Maeda K, Utsu M, Yamamoto N et al. Echogenicity of fetal lung and liver quantified by the grey-level histogram width.  Ultrasound Med Biol. 1999;  25 201-208
• 14 Maeda K, Utsu M, Kihaile P E. Quantification of sonographic echogenicity with grey-level histogram width: a clinical tissue characterization.  Ultrasound Med Biol. 1998;  24 225-234
• 15 Tekesin I, Hellmeyer L, Heller G et al. Evaluation of quantitative ultrasound tissue characterization of the cervix and cervical length in the prediction of premature delivery for patients with spontaneous preterm labor.  Am J Obstet Gynecol. 2003;  189 532-539
• 16 Tekesin I, Wallwiener D, Schmidt S. The value of quantitative ultrasound tissue characterization of the cervix and rapid fetal fibronectin in predicting preterm delivery.  J Perinat Med. 2005;  33 383-391
• 17 Tekesin I, Anderer G, Hellmeyer L et al. Assessment of fetal lung development by quantitative ultrasonic tissue characterization: a methodical study.  Prenat Diagn. 2004;  24 671-676
• 18 Streiner D L, Norman G R. Health Measurement Scales: A Practical Guide to Their Development and Use. New York: Oxford University Press; 1995
• 19 Bartko J J. The intraclass correlation coefficient as a measure of reliability.  Psychol Rep. 1966;  19 3-11
• 20 Fleiss J L, Cohen J. The equivalence of weighted kappa and intraclass correlation coeffcient as measures of reliability.  Educ Psychol Meas. 1973;  2 113-117
• 21 Khan K S, Chien P FW. Evaluation of a clinical test. I: Assessment of reliability.  Br J ObstetGynaecol. 2001;  108 562-567
• 22 Bland J M, Altman D G. Statistical methods for assessing agreement between two methods of clinical measurement.  Lancet. 1986;  1 307-310
• 23 Valentin L, Bergelin I. Intra- and interobserver reproducibility of ultrasound measurements of cervical length and width in the second and third trimesters of pregnancy.  Ultrasound Obstet Gynecol. 2002;  20 256-262
• 24 Burger M, Weber-Rössler T, Willman M. Measurement of the pregnant cervix by transvaginal sonography: an interobserver study and new standards to improve the interobserver variability.  Ultrasound Obstet Gynecol. 1997;  9 188-193
• 25 Streiner D L, Norman G R. Health Measurement Scales: A Practical Guide to Their Development and Use. New York: Oxford University Press; 1995
• 26 Kranmer M S, Finstein A R. Clinical biostatistics. The biostatistics of concordance.  Clin Pharmacol Ther. 1981;  29 111-123
• 27 Altman D G. Practical Statistics for Medical Research. London: Chapman & Hall; 1994: 403-409