Cervical Length Measured by Transvaginal Ultrasonography and Cervicovaginal Infection as Predictor of Preterm Birth Risk

Abstract

Introduction: The study shows possibilities of transvaginal sonographic measurement of the cervix in prediction of premature birth risk. Goals: The aim of the study was to follow up the cervical length in the pregnant from 16th to 37th week, as well as to do a microbiological analysis of the vaginal and cervical flora and to identify relation between the cervical shortening and microbiological flora as well as with a premature birth. Material and methods: The investigation was conducted as a prospective study on two groups of female patients in Clinical Centre of Banja Luka. In the high risk group we had 8% of patients with cervical length bellow than 15mm, 30% of patients with cervical length from 15 to 25m and 62% of patients with cervical length bigger than 25mm. In the low risk group we had no patients with cervical length bellow 15mm, 95% of patients had cervical length bigger than 25mm and 5% of patients had cervical length from 15 do 25mm. Results: The regression coefficient of the cervical length in the high risk group was 0.44mm, while in the low risk group it was 0.26mm. In the high risk group 67.56% patients had a positive cervical smear finding, while in the low risk group it was 4%. A high premature birth (defined as birth before 36.6 weeks) incidence of 50% was presented in patients with cervical length bellow 15mm. In the group of patients with cervical length up to 25mm the premature risk incidence was 10.52±0.05. In the high risk group of patients with a positive cervical smear finding, regarding the cervical length the percentage was as follows; in the subgroup of 15mm length 88,89±11,87, in subgroup from 15 to 25mm was 62,07±11,43 and in the subgroup bigger than 25mm, 60.06±8.05. Conclusion: By the analysis of the first and second goal of our study we can conclude that ultrasound assessment of cervical length is simple and feasible in the 16th week of pregnancy in both groups, with high and low risk. The length of the cervix in this period is shorter in the high risk group compared with the low-risk group. This difference was not statistically significant, however, it clearly demonstrated connection between shorten length of the cervix with the preterm birth. A regression analysis shows that the shortening of the cervix length is more frequent in high risk group which is to be expected, bearing in mind that in this group, the risk of preterm delivery was significantly higher.

128

Cervical Length Measured by Transvaginal Ultrasonography and Cervicovaginal Infection as Predictor of Preterm Birth Risk

doi: 10.5455/aim.2014.22.128-132 ACTA INFORM MED. 2014 APR 22(2): 128-132

Received: 11 December 2013 Accepted: 12 February 2014 AVICENA 2014

ORIGINAL PAPER

Cervical Length Measured by Transvaginal Ultrasonography and Cervicovaginal Infection as Predictor of Preterm Birth Risk

Arnela Ceric Banicevic1, Miroslav Popovic1, Amela Ceric2

Clinic of Gynecology and Obstetrics, Clinical Center of Banja Luka, Bosnia and Herzegovina1

Clinic of Hematology, Clinical Center of Banja Luka, Bosnia and Herzegovina2

Corresponding author: Arnela Ceric Banicevic, MD. Clinic of Gynecology and Obstetrics, Clinical Center of Banja Luka. Ul 12. Beba, 78000 Banja Luka, Bosnia and Herzegovina. Tel: [phone]; Fax:[phone]. Mob.: [phone]. E-mail: [email protected]

ABSTRACTIntroduction: The study shows possibilities of transvaginal sonographic measurement of the cervix in prediction of premature birth risk. Goals: The aim of the study was to follow up the cervical length in the pregnant from 16th to 37th week, as well as to do a microbiological analysis of the vaginal and cervical ora and to identify relation between the cervical shortening and microbiological ora as well as with a premature birth. Material and methods: The investigation was conducted as a prospective study on two groups of female patients in Clinical Centre of Banja Luka. In the high risk group we had 8% of patients with cervical length bellow than 15mm, 30% of patients with cervical length from 15 to 25m and 62% of patients with cervical length bigger than 25mm. In the low risk group we had no patients with cervical length bellow 15mm, 95% of patients had cervical length bigger than 25mm and 5% of patients had cervical length from 15 do 25mm. Results: The regression coefcient of the cervical length in the high risk group was 0.44mm, while in the low risk group it was 0.26mm. In the high risk group 67.56% patients had a positive cervical smear nding, while in the low risk group it was 4%. A high premature birth (dened as birth before 36.6 weeks) incidence of 50% was presented in patients with cervical length bellow 15mm. In the group of patients with cervical length up to 25mm the premature risk incidence was 10.520.05. In the high risk group of patients with a positive cervical smear nding, regarding the cervical length the percentage was as follows; in the subgroup of 15mm length 88,8911,87, in subgroup from 15 to 25mm was 62,0711,43 and in the subgroup bigger than 25mm, 60.068.05. Conclusion: By the analysis of the rst and second goal of our study we can conclude that ultrasound assessment of cervical length is simple and feasible in the 16th week of pregnancy in both groups, with high and low risk. The length of the cervix in this period is shorter in the high risk group compared with the low-risk group. This difference was not statistically signicant, however, it clearly demonstrated connection between shorten length of the cervix with the preterm birth. A regression analysis shows that the shortening of the cervix length is more frequent in high risk group which is to be expected, bearing in mind that in this group, the risk of preterm delivery was signicantly higher.

Key words: cervicometry, preterm birth, cervical length.

1. INTRODUCTION

Ultrasound cervicometry is a method for estimating the length of the uterus, which has its place in the prediction of late miscarriage and preterm birth. These results conrmed number of authors, and it has been proven through the rst and second predened goal of this study. In studies published the cervicometry was indicated as better method than bimanual examination in screening of high-risk groups for preterm birth (1). Ultrasonic measurement of cervical length separates a group of pregnant women with threatening preterm birth, and opens the possibility of timely responses, thus avoiding unnecessary cerclage with possible adverse consequences. As in most medical dilemmas, as well as this, there are still no standardized criteria for ultrasound nding of cervix weakening. Currently the most important parameter is the shortening of the cervix length, but the exact values still varies.

2. GOAL

The goal of the research is dened by: Estimate the length of the cervix at 16 weeks, depending on the risk group (high and low);

Monitor the length of the cervix in pregnant women in both groups (high and low risk) at 16 week to week pregnancy;

Microbiological analysis of vaginal and cervical microbial ora by taking microbiological samplesswabs, and observe any correlation between shortening of the cervix and microbial ora, and with the occurrence of preterm birth;

Evaluate the association of shortened cervix with preterm birth.

3. MATERIAL AND METHODS

This prospective study, which was conducted in the Clinic of Gynecology and Obstetrics, Clinical Center

ORIGINAL PAPER / ACTA INFORM MED. 2014 APR 22(2): 128-132

Figure 1. The dynamics of shortening the cervix length o by week pregnancy from the original value of

<15 mm measured at 16 weeks of pregnancy in high risk group

Cervical Length Measured by Transvaginal Ultrasonography and Cervicovaginal Infection as Predictor of Preterm Birth Risk

129

of Banja Luka. Part of pregnant women enrolled in the trial was hospitalized at the Department of Conservative gynecology because of various indications that did not have an impact on our testing, while part was examined as ambulatory patients at the Clinic of Gynecology and Obstetrics, Clinical Center of Banja Luka.

The respondents were divided into two groups: experimental group or a group at high risk for preterm birth and a control group or a group of low risk for preterm birth. Criteria for inclusion in the high-risk group were: pregnant women with premature birth or late miscarriage in a previous pregnancy, pregnant women with previous articial or medically indicated abortions, and women with surgeries of the cervix, in pregnant women with suspected or proven cervical infections.

4. RESULTS

In the period from January 1st 2007 to September 1st 2010 the study included 200 women. They are selected by criteria dened as a high-risk group (n=100), and a low-risk group (n=100). Each group began with participation in the study after taking the anamnesis at rst examination in 16th week of pregnancy. It included classic gynecological examination, transvaginal ultrasound examination with measurement of cervical length, taking a cervical smear, and fetal biometry with routine laboratory tests as dened by the protocol.

Group Length of the

cervix % S%

High risk Less than 15 mm 8.0 0.02 From 15 to 25 mm 30.0 0.04 Longer than 25mm 62.0 0.04

Low risk Less than 15 mm - - - From 15 to 25 mm 5.0 0.06 Longer than 25mm 95.0 0.02

Table 1. Relative frequency of subjects observed by analytical subgroups with dierent lengths of cervix in high-risk and low-risk groups for preterm birth

Risk group High risk (%) Low risk (%) Positive nding 67.56 7.80 4.00 1.95 texp.=7.901 t0.05 = 1.667t0.01=2.381

Table 2. Relative frequency of positive cervical smear in both groups

Cervix length Mean (%) Standard error Less than 15 mm 88.89 11.87

From 15 to 25 mm 62.07 11.43 Longer than 25 mm 60.06 8.05

Table 3. Relative frequency of positive cervical smears according to the length of the cervix in the high risk group and its statistical signicance

Isolated pathogen %(n)

Enterococcus 42.85(27)E. Coli 20.63(13) Streptococcus 20.63(13) Gardnerella 6.39(4) Staphylococcus hom. 3.17(2) Staphylococcus ep. 3.17(2) Proteus mirabilis 1.58(1) Ureaplazma urealiticum 1.58(1)

Table 4. The frequency of isolated pathogens in cervical smears in high risk group

Isolated pathogen 16 week 36.6 week

16w-36.6w in mm

Gardnerella 32.65 2.25 21.40 2.78 11.25E. Coli 33.44 1.95 23.15 1.27 10.29 Enterococcus 31.21 2.01 21.42 1.52 9.79 Streptococcus 29.97 4.39 20.89 2.59 9.08

Table 5. The dierence in the length of the cervix according to the cause of infection in the 16th and 36.6 week of pregnancy

Isolated pathogen % of preterm birth Enterococcus 37.5 Staphylococcus 12.5 Streptococcus 12.5E.Coli 12.5 Without infection 25.0

Table 6. The incidence of preterm birth before 37 weeks depending on the isolated pathogen from cervical swabs in the high risk group.

Cervix length

Dynamics of cervical length by week of pregnancy

18

16

y = -0.4075x + 21.774 R = 0.9652

Cervical length in mm

14

12

10

8

6

4

2

0 16 18 20 22 24 26 28 30 32 34 36 38 40

Weeks of pregnancy

Figure 1. The dynamics of shortening the cervix length o by week pregnancy from the original value of <15 mm measured at 16 weeks of pregnancy in high risk group

Table 2. Relative frequency of positive cervical smear in both groups

Risk group High risk (%) Low risk (%)

Positive finding 67.56 7.80 4.00 1.95

t =7.901 t = 1.667 t =2.381

Preterm delivery inweek34-34.6 35-35.6 36-36.6 % % % <15mm 25 25 50 15-25mm - - 100 >25mm - - 100

Table 7. Relative frequency of preterm birth in certain weeks, depending on the length of the cervix in high risk group

ACTA INFORM MED. 2014 APR 22(2): 128-132 / ORIGINAL PAPER

130

Cervical Length Measured by Transvaginal Ultrasonography and Cervicovaginal Infection as Predictor of Preterm Birth Risk

Respondents X Sx t exp

Cervix length up to 25mm 75.00 12.31 2.872 Cervix length above 25mm 25.00 12.31

t 0.05=1.894 t 0.01=2.998

Table 8. Relative frequency of preterm birth, depending on the length of the cervix in high risk group

Respondents in the high risk group

Preterm delivery (XSx)

t exp

Microbiological swabpositive 87.50 11.69 4.537*

Microbiological swabnegative 12.50 11.69

t 0.05=2.920 t 0.01=6.965

Table 9. Relative frequency of preterm birth, depending on the ndings of a positive cervical smear in the high risk group

5. DISCUSSION

In this study we included two groups of pregnant women, a group at high risk for preterm birth (n=100) and a group of low risk for preterm birth (n=100). In the high risk group in 16th week short cervix had 8% of pregnant women, 30% of the truncated normal pregnant women and 62% of them. In the low-risk group 95% of pregnant women had normal cervical length and truncated was of 5%.

Mean values of the cervix length at the high-risk group were 30.8 mm and in the low-risk group 35.0 mm. Thus, the length of the cervix in the high risk group was shorter than it was in the low-risk group. It is expected that the data is rst clearly dened in our study, because we know that a history of risk is very important in determining and categorizing the general population of pregnant women with regard to the risk of preterm birth. From our results it is clear that in the high-risk group, dened by previously determined parameters, there is clearly a shorter length of the closed portion of the cervical canal and the length of the cervical canal shows a history of previous miscarriage and/or preterm labor in previous pregnancies is an important fact and risk factor for the next pregnancy. This is a very important observation that has not been clearly dened in other clinical trials and certainly opens up new possibilities for the preparation and design of future studies that will address cervicometry as a method of detection risk group for preterm birth. Dynamics of shortening the door was followed from the 16 to 37 weeks of pregnancy, which was the second goal of our research. The largest regression coefficient in this study were obtained by the group of high-risk long neck above 25mm and he was 0.65 mm, followed by a group of cervix length below 15mm 0.41 and a group of 15 to 25mm where the regression coefficient was 0.27 mm per week . In the group of low-risk weekly regression coefficient was 0.26 mm. These data are also original and very interesting because they were not so far the subject of analysis. In our results is clear that the regression coefficient was higher in the high risk group

than in the low-risk group. It is to be expected because bearing in mind the relatively high average value of cervical length at baseline (week 16) , and the connection between the length of the cervix with preterm birth in the high risk group was expected that there must be a signicant shortening of the length of the uterus cervix during pregnancy. It was proved in our study that the largest regression coefficient was in the high risk group of pregnant women who have had the longest cervix at 16 week. And this fact with its originality will have an eect on the design of future studies as well as new screening programs screening of pregnant women at risk for preterm birth.

Our results are consistent with the published studies. The study of Hee Lee Yoong and associates, 2009, in the group of low risk for preterm birth cervix length was measured after 22 week and was 35.70 mm, and at 28 week 33.70 mm. There was no signi-cant regression in the length of the cervix (2). And it is relatively expected data, as low risk populations who have not observed an increased incidence of pre-term birth and there is no reason for the signicantly greater shortening of cervical length of 22 to 28 weeks of pregnancy, and it is expected that by symbolically shortening should come later. In the high risk group cervix length in 15 week amounted to 36mm, in the 20 week 35 mm, and in the 25 week 33.80 mm. In this study has proven that ultrasonic measurement of the cervix length between 16 and 24 week has signicance.

Berghella et al. Dene ultrasound as a method for assessment of the length of the door, or for the prediction of preterm delivery (3). They pay special attention to proper technique of performance, which is essential for accurate results. Ultrasonic measurement of cervical length in comparison with clinical bimanual examination showed a better diagnostic value to assess the risk of preterm birth in the general population of pregnant women. A classic digital gynecological examination in pregnancy also have been traditionally used in everyday practice and has not proved as eective in detecting cervical weakness, because often the exterior mobility of the cervical canal is wrongly declares pathological condition. There is increasing evidence that the cervicometry can provide very valuable information in support of the weakness of the cervix.

Matijevic et al. conducted randomized controlled study, which followed the diagnostic accuracy of sonographic assessment of cervical length and clinical examination in the second quarter (4). Total of 282 patients, asymptomatic, singleton pregnancy. Experimental group (n=138) and control group (n=144). Shortened cervix was dened as 24 mm, 5 percentile. In the control group, Bishop score 4 is dened as high. Preterm labor is dened as every before 37 weeks. Shortened cervix was found in 6/138 patients (4.3%). Bishop high score in 17/144 patients (11.8 %). The incidence of preterm birth 16/282 patients or 5.7 %. Sensitivity of 57.1 compared to 33.3, the positive

ORIGINAL PAPER / ACTA INFORM MED. 2014 APR 22(2): 128-132

Cervical Length Measured by Transvaginal Ultrasonography and Cervicovaginal Infection as Predictor of Preterm Birth Risk

131

predictive value of 66.7 % compared to 17.6 %. Shortened cervix in relation to the high Bishop score had a 12 times greater chance of preterm birth in low risk group. The conclusion is that ultrasound evaluation has better diagnostic accuracy in predicting preterm birth in relation to digital examination in low-risk group.

In this study, we analyzed the microbial vaginal/ cervical ora in both groups, with high and low risk, and the observed correlation between length of the cervix and positive microbial ora with premature birth. In the group of high-risk we had a positive microbiological nding of cervical smear in 67.56 % of patients, while in the low-risk group we had only 4 % of positive ndings. In the group with high-risk, short cervix <15mm , positive results of microbiological swabs had 88.89 % of patients , with the shorter cervix length of 15 to 25mm positive swab had 62.07 % and 60.06 % of patients in the group with normal length of the cervix above 25mm. Ttest of signicance showed a statistically signicant higher relative frequency of positive cervical smear ndings in the high risk group to low-risk and high-risk groups within the signicantly higher relative frequency with cervix lengths below 15mm. This is very important information because it connects the positive microbial ora and the length of the cervix and connecting the two etiological factors of preterm birth infection and cervical insufficiency. It is not possible to determine which of these two is more important, and/or which appeared before and perhaps beneted to another. It is our opinion that the infection is still what it was favorable for shortening the length of the cervix, as we observed shorter length of the cervix in patients with a positive swab. Also, there needs to bear in mind the fact that the positive results of cervical smear was found before inclusion in the study, or more than 50 % of the women groups at high risk. We tracked the incidence of preterm birth in both groups of patients with positive microbiological ndings of cervical smears. In the group of high-risk with cervix length up to 15mm 25% of patients had a positive cervical smear and the birth was after 34week, 25 % in 35 week and 50 % in 36 week. Patients with shortened cervix length with shorter cervix in high-risk group with positive cervical smear positive had preterm births in 36 week. When we analyzed the relative frequency of preterm birth according to the ndings of positive microbiological ndings of cervical smears, we noticed that 87.50 % had a positive smear and preterm birth, or 12.5% negative cervical smear.

Matijevic et al., In the study of bacterial vaginosis and cervical infections, for prediction of preterm labor, dened any before 37 weeks, and early pre-term birth 34 weeks in the group of low risk n=316 (5). Vaginal pH was measured by the test glove and cervical length (CL) by transvaginal ultrasound. PH value measured less than 5.0 (95th percentile) and CL larger than 26mm (5th percentile) were consid-

ered normal ndings. Of the 14 patients (4.4%) were with higher vaginal pH, and 15 (4.7%) with a shortened length of the cervix. The incidence of preterm birth was 7.2%, while the incidence of preterm birth was 2.5%. The length of the door in the study was signicantly correlated with preterm birth but not with early preterm birth. Elevated vaginal pH was a better predictor of early preterm birth of a shortened neck in the group of low risk.

In this study, we followed two groups, a group at high risk with previous preterm births, previous abortions, previous interventions on the cervix, shortening of the cervix and positive microbiological ndings of cervical smears. In the low-risk group, were patients that did not have load of previous pregnancies and the proper length of the cervix. In the high risk group controls were made in dependence on the cervix length, so for the patients with shortest cervix < 15 mm we measured by ultrasound cervix length each week, with a length of 15 to 25mm every other week, and with cervix length > 25mm as in the group of low risk every 4 weeks. In the study, we concluded that a group with cervix length below 15 mm had a 50 % incidence of premature births, which was statistically signicant compared to the length of the cervix above the 25 mm in high-risk group. If we display cervix length of 25mm as the normal length, and a group of up to 25mm as pathological ndings, 75 % of the total number of premature births would be in this group. In this way, by the study we have proved that a group of high-risk patients with short cervix in the 16th week is the one that needs to be monitored as a high risk for preterm birth.

Study of Owen et al. speaks about length of the cervix that has a 3.3 times higher relative risk of pre-term delivery (7). The study lasted from March 1997 until November 1999 at nine university medical centers in the United States. Included were 183 patients, singleton pregnancy, which used to have a spontaneous preterm birth before 32 weeks of gestation. The rst review was carried out between 14 and 18 weeks of gestation. Sonographic measurement was carried out in two-week intervals. It is also accompanied by the so-called funneling and dynamic cervical shortening. Preterm labor is dened as the one before 35 weeks of gestation. Of the 138 patients, 48 (2%) had a preterm birth before 35 weeks. Cervical length < 25mm at the rst sonographic measurement is associated with a relative risk of spontaneous preterm birth of 3.3 (95% condence interval, sensitivity 19 %, spec-icity 98 % and positive predictive value 75%). The study concluded that ultrasonographic monitoring of cervical length between 16 and 18 weeks as serial assessments at biweekly intervals has signicant predictive value for preterm birth before 35 weeks.

Grgic et al. followed the length of the cervix and gland parameters (QGCS), ultrasound measurement in the second quarter of predicting preterm birth in low-risk populations (81). The prospective study included 278 patients without symptoms of preterm

ACTA INFORM MED. 2014 APR 22(2): 128-132 / ORIGINAL PAPER

132

Cervical Length Measured by Transvaginal Ultrasonography and Cervicovaginal Infection as Predictor of Preterm Birth Risk

labor, nulliparous, singleton pregnancies. CL and QGCS were monitored from 16 to 23 weeks of pregnancy. CL 24 mm was dened as shortened. QGCS parameters were: cervical mucus thickens and invasion of cervical glands. Score 1 is dened as low. Shortened cervix was present in 6.1 (17/278) and low QGCS in 5.7 % (16/278). The incidence of preterm birth before 34 weeks was 2.4% (6/278) and between 34 and 37 week 3.2% (9/278). QGCS low score compared to the cervix was superior in sensitivity (83.3 % vs. 66.6%) for preterm delivery before 34 weeks, sensitivity 55.5% vs. 22.2%. The positive predictive value was 31.2% vs. 23.5% for delivery before 34 week and 31.2% vs. 11.7% for delivery between 34 and 37 weeks. The conclusion was that QGCS has the same if not better predictive value compared with shortened CL for prediction of preterm birth.

Finally, we conclude that we have through our results and the results of other authors in this study failed to show a clear connection between a shortened cervix with preterm birth. Accordingly, cervicometry seems to be one of the best methods available to us in the diagnosis and dening risk groups for preterm birth. In addition, infection as an etiological factor should not be neglected even independently of other screening methods as in our study, lack of association with infection and cervical insufficiency, and the combination of these two risk factors and clearly dominant becomes associated with preterm birth in both study groups.

6. CONCLUSION

1. The analysis of the rst and second goal of our study we can conclude that ultrasound assessment of cervical length is simple and feasible in the 16th week of pregnancy in both groups, with high and low risk.

2. The length of the cervix in this period is shorter in the high risk group compared with the low-risk group. This difference was not statistically signicant, however, it clearly demonstrated connection between shorter length of the cervix with the preterm birth.

3. A regression analysis shows that the shortening of the length of the cervix was larger in high risk group which is to be expected, bearing in mind that in this group, the risk of preterm delivery was signicantly higher.

4. Microbiological analysis of cervicovaginal smear is also a very useful data for determining risk groups for preterm birth. Pregnant women with positive ndings have much higher incidence of preterm birth than those with normal vaginal ora.

5. Most frequent was infection with G. vaginalis, as one of the dominant causes of bacterial vaginosis in which exist emphasizes risk due to bacterial vaginosis as a signicant cause of premature birth. It is also important a connection between infections and shortened cervix as a cause of preterm birth, but unfortunately

these results was not possible to determine which of these two factors is more important, and was primarily present.

CONFLICT OF INT EREST: NONE DECLARED.

REFERENCES

6. Draancic A i sur. Porodnitvo. kolska knjiga Zagreb 1994; 223-232.

7. World Helath Organisation : Prevention of perinatal mortalitiy. Public Health Papers 42 Geneve.

8. American Academy of Pediatrics Committee on fetus and newborn : Nomenclature for duration of gestation, birth weight, and intrauterine growth. Pediatrics. 1967; 39: 935-939.

9. Henon MA, Katz M, Creasy RK. Evaluation of preterm birth prevention program: preliminary report. Obstet Gynecol. 1982; 59: 452-456.

10. Grgic O, Matijevic R. Dilemmas about antenatal use of corticoksteroids for prevention of neonatal morbidity and mortality. Acta Med Croatica. 2005; 59(2): 129-135.

11. Kurjak A. i sur. Ultrazvuk u ginekologiji i porodnitvu. Art studio, Zagreb, 2000: 322-330.

12. Draancic A. Osjetljivost i aktivnost uterusa na kraju trudnoce. Disertacija. Medicinski Fakultet Sveucilita u Zagrebu, Zagreb 1964.

13. Hillier K, Wallis P. Prostaglandins, steroids and human cervix. U:: Ellwood DA, Anderson AM (eds). The cervix in pregnancy and labor. Clinical and biochemical investigation. Edinburgh, Churchill Livingstone 1981; 34-40.

14. Gire C, Faggianelli P, Nicaise C, Shojai R, Fiori A, Chan C, Boubli L, D Ercole C. Ultrasonographic evaluation of cervical length in pregnancies complicated by pre-term premature rupture of membranes. Ultrasound Obstet Gynecol. 2002; 19: 565-569.

15. Kuvacic I, krablin S, Hodzic D, Milkovic G. Effect of expatriation on the outcome of pregnancy. Arch Mat Child Health. 1992; 36: 143-54.

16. Matijevic R, Grgic O. Clinical examination and transvaginal sonography in mid trimester as potencial screening tests for preterm labor. Preliminary results on low risk population. J Matern Fetal Neonatal Med. 2004; 16: 48.

17. Matijevic R, Grgic O. Ritodrin in oral maintenance of tocolys after active preterm labor: randomized controlled trial. Croat Med J. 2006; 47(1): 25-31.

18. Matijevic R,Grgic O, Klobucar A. Diagnosis and therapy of cervical insufciency. Acta Med Croatica. 2003; 57(4): 287-294.

19. Barden TP. Premature labor. U: Pitkin RM, Zlatkin FJ. Yearbook of obstetrics and Gynecology Chicago : Year book medical publishers, 1977: 109.

20. Fuchs F, Husslein P. Prevention and management of prematurity. U: Lauersen NH, ur. Management of high risk pregnancy. New York: Plenum medical group company, 1983; 375-395.

ORIGINAL PAPER / ACTA INFORM MED. 2014 APR 22(2): 128-132