1 Department of Obstetrics and Gynecology, and 2 Department of Hygiene, Ioannina University School of Medicine, 45500 Ioannina, Greece and 3 Department of Obstetrics and Gynecology, Bradford Royal Infirmary, Duckworth Lane, Bradford, West Yorkshire, UK
4 To whom correspondence should be addressed. Email: kaponisapostolos{at}hotmail.com
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: carbon dioxide/distending media/normal saline/out-patient hysteroscopy
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In the international literature, there are several studies comparing the use of CO2 with normal saline in hysteroscopy (Soderstorm, 1992; Goldfarb, 1996
; Nagele et al., 1996b
; Perez-Medina et al., 2000
; Brusco et al., 2003
; Litta et al., 2003
; Pellicano et al., 2003
). These studies mainly refer to the hysteroscopic technique, the mode of administration and the tolerance of the patients in relation to the distending medium. Perez-Medina et al. (2000)
prefer the use of normal saline because of the ability to perform minor procedures. In contrast, Soderstorm (1992)
prefers the use of CO2 in diagnostic hysteroscopy. Nagele et al. (1996b)
have reported that there were no significant differences in the hysteroscopic image regarding the use of gas or liquid as a medium for distension, although there was a more frequent occurrence of gas bubbles with the use of CO2. Goldfarb (1996)
reported that during diagnostic hysteroscopy with CO2, the presence of submucous myomas generated difficulties for the determination of the percentage of their mass, which appeared in the endometrial cavity, as there was inadequate distension and a large amount of mucus, bubbling and bleeding. Therefore, there was a more accurate diagnosis of myomas when normal saline was used. In more recent studies (Brusco et al., 2003
; Litta et al., 2003
; Pellicano et al., 2003
), although the authors were in agreement that both media offer high quality images, the results tend to suggest that the administration of normal saline offers more benefits in out-patient hysteroscopy. These studies supported that normal saline is more acceptable to patients, reduces the time of the procedure and requires fewer auxiliary procedures, such as cervical canal dilation or administration of cervical anaesthesia.
With the exception of the study by Goldfarb (1996), where there is reference to the presence of myomas, to our knowledge, there is no other publication in the international literature relating the distending media to diagnostic accuracy of out-patient hysteroscopy. A common feature of all the above-mentioned studies is that no method of verification of hysteroscopic findings was applied and, therefore, the results presented are based exclusively on the experience of the physicians. For this reason, a prospective, randomized study was designed having as objective the evaluation of the diagnostic accuracy of hysteroscopy on an out-patient basis, in relation to the medium used for distension. In addition, patient's discomfort after out-patient hysteroscopy with normal saline or CO2 was measured.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
Total abdominal hysterectomy was performed 1224 h after the hysteroscopy and the surgical specimen was submitted immediately to the Pathology Laboratory of Ioannina University School of Medicine for histological examination. The specimen was dissected directly, by two pathologists who had been specifically instructed regarding the need to detail findings about the uterine cavity, the macroscopic findings were recorded and afterwards the samples were prepared for histological examination. The macroscopic findings and the diagnosis of the histological examination were related to the hysteroscopic diagnosis. Only cases with submucous myomata (type 0, type I and type II) were taken into consideration according to the classification system of the European Society of Hysteroscopy. The histological diagnosis served to verify the hysteroscopic diagnosis.
Study population
The population of this study comprised 74 women (see Figure 1; flow diagram). There were no differences between groups in terms of age, parity and menopausal status (Table I). Among post-menopausal patients, two women of group B were receiving HRT. The mean age of women in both groups was 52.8 years (range 3978). Within group A, the mean age of women was 52.3 years (range 3978) and in group B it was 53.5 years (range 4471). The main reason for the admission of women to our Department was abnormal uterine bleeding (menorrhagia, polymenorrhoea or post-menstrual bleeding) (72 women), whereas in two cases the hysterectomy was performed due to chronic pelvic pain. Within the framework of pre-surgical examination, all women underwent laboratory testing, clinical examination of the pelvis and transvaginal ultrasonography. All women were fully informed of all aspects of the study. Full informed consent for the study, which was approved by the local medical ethics committee, was obtained from all subjects.
|
The flow and the pressure of the medium for distension were kept at the lowest possible level required for the completion of the hysteroscopy. In group A (CO2), the hysteroscopy started with a gas flow of 25 ml/min and a pressure of 40 mmHg. In many cases, this pressure was inadequate for panoramic examination of the endometrial cavity and the imaging of the tubal ostia, and therefore it was increased during the examination. The maximum pressure, which was used in order to distend the endometrial cavity, was 100 mmHg. Normal saline perfusion (group B) was performed with gravity-controlled pressure (de Wit et al., 2003). A container with a capacity of 1 l was raised 1 m above the level of the woman. A pressure cuff was connected with a manometer surrounding the container. The initial pressure applied to the container was 40 mmHg. In some cases, due to an inadequate view, this pressure was increased up to 80 mmHg (Cooper and Brady, 2000
).
Equipment
All hysteroscopies were performed with the same equipment, regardless of the distending medium. The Bettochi hysteroscope (Karl Storz, Tuttlingen, Germany) having a diagnostic sheath with a diameter of 2.8 mm and lances with an inclination of 30° was used. The Hamou microhysteroflator with a maximum pressure of 200 mmHg and a gas flow of 25 ml/min was used for the infusion of CO2. The illumination of the endometrial cavity was achieved by a source supplying cold light xenon illumination, Olympus CLV-S20, through a wire of optical fibres (Olympus, AT30/FF-IF11C5). We used the Olympus OTV-S5 endocamera that transformed the signal from the CCD into a video signal for projection onto a screen. The Olympus DEV202 20-inch screen monitor reproduced images of high precision. All of the hysteroscopies were recorded on video.
Analysis of the results
Student t-test was used for the comparison of the prevalence and intensity of symptoms between the two groups. The 2 test with Yates' correction was used to control the homogeneity between groups A and B. A P-value <0.05 was considered to be statistically significant. The sensitivity, specificity, positive predictive value, negative predictive value and the diagnostic accuracy of each hysteroscopic method (group A, CO2; group B, NaCl) were estimated. The receiver operating characteristic (ROC) curves were designed for any isolated finding of the endometrial cavity. The program SPSS 10.1 (Statistical Package for the Social Sciences) was used for the design of the ROC curves.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Prevalence of pathological endometrial findings
The indication to perform total abdominal hysterectomy was abnormal uterine bleeding in 72 women and chronic pelvic pain in two women. 2 test (Yates' correction) did not demonstrate any significant difference between the two populations of group A and group B (P>0.05) regarding the prevalence of endometrial pathology (i.e. the proportion of true submucous myomas, proportion of true polyps, etc.). Thus, the two groups were comparable. Myomas were observed in 28 (71%) patients of the CO2 group compared with 22 (63%) patients of the normal saline group, whereas comparative respective values for polyps were 18 (46%) versus 20 (57%), for hyperplasia 15 (38%) versus 21 (60%), and for subtle endometrial lesions such as small polyps, marked and moderate mucosal elevations, crypts and endometrial hypervascularization 23 (59%) versus 24 (68%) (Table II). The latter findings were most often indicative for adenomyosis or endometritis, according to the histological diagnosis. Specifically, for adenomyosis, hysteroscopic indicative findings were the image of crypts and/or the presence of thick, dilated endometrial vessels (Ota and Tanaka, 2003
). Indicative hysteroscopic findings for endometritis were irregular, swollen glands that stand out on a reddish hypotrophic endometrium, the so-called strawberry pattern (Paavonen et al., 1985
; Thomas and Michels, 1995
). The presence of these findings makes the endoscopist suspicious for adenomyosis or endometritis.
|
|
|
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In the current study, we evaluated the diagnostic accuracy offered by the two most used distension media (CO2 and normal saline) for out-patient hysteroscopy. The hysteroscopic findings were compared with the diagnosis by histological examination after hysterectomy, performed 1224 h, post-hysteroscopy. Hysterectomy specimens are regarded as the criterion standard for verification of endometrial disease (Clark et al., 2002) and subsequently for the hysteroscopic findings. Earlier studies attempted to evaluate the diagnostic accuracy of hysteroscopy in relation to the medium used for distension (Soderstorm, 1992
; Goldfarb, 1996
; Nagele et al., 1996b
; Perez-Medina et al., 2000
; Brusco et al., 2003
; Litta et al., 2003
; Pellicano et al., 2003
). Their results were based exclusively on the experience of the hysteroscopist without any verification of the hysterocopic findings. Our study is novel because, by using the macroscopic observation of the surgical specimen and the histological examination, the hysteroscopic findings are confirmed. The above-mentioned studies are restricted to the comparison between hysteroscopic techniques and between the quality of the image without evaluating the diagnostic accuracy of each method.
After evaluation of the results, we can suggest that, regardless of the distending medium, hysteroscopy offers high sensitivity, specificity and diagnostic accuracy concerning the examination of the endometrial cavity and represents the most accurate diagnostic procedure to monitor intra-uterine lesions. In the current study, the differences in diagnostic accuracy between the two methods, as also shown in the ROC curves (Table IV and Figure 3), were not significant for the detection of major pathology of the endometrial cavity. Normal saline allows the detection of submucous myomas, large polyps and hyperplasia with slightly higher diagnostic accuracy compared with CO2, although the difference was not significant. In these cases, the presence of blood in the endometrial cavity and the inability of gas to remove it was the main reason that hysteroscopy with normal saline provided greater diagnostic accuracy. The latter hypothesis has already been supported by other authors (Goldfarb, 1996; Nagele et al., 1996b
; among others). In contrast, a statistically significant difference was observed between the two methods of hysteroscopy (use of CO2 versus use of NaCl) regarding the detection of subtle endometrial lesions (usually corresponding to the histological diagnosis of adenomyosis or endometritis). Campo et al. (1999)
have also reported that the use of saline as a distension medium improves the diagnostic capacity for subtle endometrial lesions. In the current study, the use of CO2 had lower diagnostic accuracy (64.6%) and sensitivity (0.30) in comparison with the cases where normal saline was used for the distension of the endometrial cavity (diagnostic accuracy 85.4%, sensitivity 0.79). This is attributed to the stable direction of pressure from the gas on the endometrial walls. As a result, subtle endometrial lesions (small polyps, mucosal elevations, crypts and hypervascularization) which usually can be observed hysteroscopically in cases of adenomyosis and/or endometrial inflammations were immersed in the normal surrounding endometrium and were not visible during the hysteroscopy. On the other hand, during hysteroscopy with normal saline, the pressure of the distending medium on the endometrial walls did not have the same direction during the hysteroscopic examination since the fluid was leaking out of the cervical os. As a result, these subtle lesions seemed to float inside the distending medium and the endoscopist has the ability to observe them.
Abnormal uterine bleeding was the main reason for the admission of women in our Department. These haemorrhages were due mainly to the presence of myomas (67.5%; mean prevalence of myomas of both groups). There was also an increase in the prevalence of endometrial hyperplasia (48.6%; mean prevalence of hyperplasia of both groups). These facts must be attributed to the selection of the specific population as the mean age was quite high and all of the women were admitted in order to undergo hysterectomy. When normal saline was used as the distending medium, hysteroscopy was completed successfully in a higher percentage of women compared with CO2. However, this fact cannot be attributed to the distending media. The intense symptoms of the vagal reaction which appeared in one woman during hysteroscopy with CO2 were probably caused by the idiosyncrasy of the patient rather than the medium for distention.
Previous studies (Nagele et al., 1996b; Brusco et al., 2003
; Pellicano et al., 2003
) have suggested that the use of CO2 in out-patient hysteroscopy causes more intense symptoms for the patients in comparison with the use of normal saline. In contrast, the study of Litta et al. (2003)
supported that normal saline causes more intensive discomfort than CO2. In the present study, the great majority of patients of both groups did not feel any intense pain. This may be attributed to (i) the hysteroscopic method (vaginoscopic approach) that was applied (Paschopoulos et al., 1997
; Okeahialam et al., 2001
) and (ii) the administration of the distending media with a lower flow and pressure. We can suggest that the distending medium did not influence the tolerance of the procedure by women.
During hysteroscopy we did not attempt to perform endometrial biopsy or any other operation because of the impending hysterectomy (1224 h later). The aim was to avoid any changes in the appearance of the endometrial cavity and, therefore, the endoscopist and the pathologist (during the macroscopic examination of the specimen) had the same objective findings for diagnosis.
A limitation of the current study was the small sample size. This sample size was adequate for the detection of a statistically significant difference in the diagnostic accuracy among hysteroscopy with normal saline and hysteroscopy with CO2 in cases of subtle endometrial lesions. However, in all other cases of endometrial pathology, a study population 510 times higher will be needed to detect any significant difference in the diagnostic accuracy of the two hysteroscopic methods (based on power calculations from the results of the current study; Altman, 1991). Another limitation of the current study was that more than half of the women comprising the study population were post-menopausal and, therefore, our results may not be generalizable to women of reproductive age or asymptomatic women with infertility.
Based on these results, we can suggest that both distending media were comparable regarding patient discomfort. In addition, CO2 and normal saline presented with similar high diagnostic accuracy in cases of submucous myomas, polyps or hyperplasia. In contrast, normal saline seems to be the most appropriate medium for the detection of subtle endometrial lesions.
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Brusco GF, Arena S and Angelini A (2003) Use of carbon dioxide versus normal saline for diagnostic hysteroscopy. Fertil Steril 79, 993997.[CrossRef][ISI][Medline]
Campo R, Van Belle Y, Rombauts L, Brosens I and Gordts S (1999) Office mini-hysterocopy. Hum Reprod Update 5, 7381.
Clark TJ, Voit D, Gupta JK, Hyde C, Song F and Khan KS (2002) Accuracy of hysteroscopy in the diagnosis of endometrial cancer and hyperplasia. J Am Med Assoc 288, 16101621.
Cooper JM and Brady RM (2000) Intraoperative and early postoperative complications of operative hysteroscopy. Obstet Gynecol Clin North Am 27, 347366.[ISI][Medline]
de Wit AC, Vleugels PH and de Kruif JH (2003) Diagnostic hysteroscopy: a valuable diagnostic tool in the diagnosis of structural intra-cavital pathology and endometrial hyperplasia or carcinoma? Six years of experience with non-clinical diagnostic hysteroscopy. Eur J Obstet Gynecol Reprod Biol 110, 7982.[CrossRef][ISI][Medline]
Goldfarb HA (1996) Comparisons of carbon dioxide with continuous-flow technique for office hysteroscopy. J Am Assoc Gynecol Laparosc 3, 571574.[ISI][Medline]
Litta P, Bonora M, Pozzan C, Merlin F, Sacco G, Fracas M, Capobianco G and Dessole S (2003) Carbon dioxide versus normal saline in outpatient hysteroscopy. Hum Reprod 18, 24462449.
Loffer F (1995) Complications of hysteroscopy, their cause prevention and correction. J Am Assoc Gynecol Laparosc 3, 1126.[ISI][Medline]
Nagele F, O'Connor H, Davies A, Badawy A, Mohamed H and Magos A (1996a) 2500 outpatient hysteroscopies. Obstet Gynecol 88, 8792.
Nagele F, Bournas N, O'Connor H, Broadbent M, Richardson R and Magos A (1996b) Comparison of carbon dioxide and normal saline for uterine distention in outpatient hysteroscopy. Fertil Steril 65, 305309.[ISI][Medline]
Okeahialam MG, Jones SE and O'Donovan P (2001) Early experience with the vaginoscopic approach to outpatient hysteroscopy. Gynecol Endosc 10, 5759.[CrossRef]
Ota H and Tanaka T (2003) Stromal vascularization in the endometrium during adenomyosis. Microsc Res Tech 60, 445449.[CrossRef][ISI][Medline]
Paavonen J, Kiviat N, Brunham RC, Stevens CE, Kuo CC, Stamm WE, Miettinen A, Soules M, Eschenbach DA and Holmes KK (1985) Prevalence and manifestations of endometritis among women with cervicitis. Am J Obstet Gynecol 152, 280286.[ISI][Medline]
Paschopoulos M, Paraskevaidis E, Stefanidis K, Kofinas G and Lolis D (1997) Vaginoscopic approach to outpatient hysteroscopy. J Am Assoc Gynecol Laparosc 4, 465467.[ISI][Medline]
Pellicano M, Guida M, Zullo F, Lavitola G, Cirillo D and Nappi C (2003) Carbon dioxide versus normal saline as a uterine distention medium for diagnostic vaginoscopic hysteroscopy in infertile patients: a prospective, randomized, multicenter study. Fertil Steril 79, 418421.[CrossRef][ISI][Medline]
Pellicer A and Diamond M (1988) Distenting media for hysteroscopy. Obstet Gynecol Clin North Am 15, 2328.[ISI][Medline]
Perez-Medina T, Bajo JM, Martinez-Cortes L, Castellanos P and Perez de Avila I (2000) Six thousand office diagnostic-operative hysteroscopies. Int J Gynecol Obstet 71, 3338.[CrossRef][ISI][Medline]
Siegler A (1975) A comparison of gas and liquid for hysteroscopy. J Reprod Med 15, 7375.[ISI][Medline]
Soderstorm R (1992) Distending the uterus: what medium is best? Clin Obstet Gynecol 35, 2528.
Thomas C and Michels MPH (1995) Chronic endometritis. Am Fam Physician 52, 217222.[ISI][Medline]
Submitted on March 10, 2004; resubmitted on May 24, 2004; accepted on July 20, 2004.
|