Carbon dioxide versus normal saline in outpatient hysteroscopy

Pietro Litta1, Michela Bonora1, Chiara Pozzan1, Federica Merlin1, Giuseppe Sacco1, Mara Fracas1, Giampiero Capobianco2 and Salvatore Dessole2,3

1 Department of Gynaecology and Obstetrics, University of Padua and 2 Department of Pharmacology, Gynaecology and Obstetrics, University of Sassari, Italy

3 To whom correspondence should be addressed at: Department of Pharmacology, Gynaecology and Obstetrics, University of Sassari, Viale San Pietro 12, Sassari 07100, Italy. e-mail: dessole{at}uniss.it


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: The aim of this prospective randomized study was to measure patients’ discomfort after hysteroscopy with CO2 or normal saline. METHODS: A total of 415 patients was randomized to two groups according to distension medium (CO2, n = 201; normal saline, n = 214). The nature of randomization was to alternate distension media on a weekly basis. After hysteroscopy, women were asked to rate the pain experienced on a 100-mm visual analogue scale (0 = no pain; 100 = worst imaginable pain). Pain scores were expressed as mean ± SD (0–40 = minimal; 41–70 = moderate; 71–100 = severe). Data were analysed using Student’s t-test. RESULTS: Irrespective of the distension medium used, pelvic discomfort was worse in nulliparous women (pain score 39.0 ± 26.5) than in multiparous women (30.4 ± 25.9) (P < 0.05), especially if they were premenopausal. For all patients and both distension media, pelvic discomfort was generally minimal but higher in patients who had undergone hysteroscopy with normal saline (P < 0.05). CONCLUSIONS: CO2 and normal saline were comparable with regard to patient discomfort, but in terms of the high frequency of abnormal uterine bleeding, normal saline may be the most appropriate distension medium for outpatient hysteroscopy.

Key words: distension media/outpatient hysteroscopy/pain measurement/visual analogue scale


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Nowadays, hysteroscopy is the ‘gold-standard’ procedure used to describe the morphology of the uterine cavity and the presence of intrauterine lesions (De Iaco et al., 2000Go). The success of this procedure depends on the instruments used, the operator, and the distension medium (Baggish, 1989Go). Direct visualization of the uterine cavity allows the detection of organic lesions such as polyps and submucosal myomas, and directed biopsy can also be performed (Vercellini et al., 1994Go).

Hysteroscopy requires distension of the uterine cavity with either a gas (CO2) or liquid medium (Hyskon; normal saline, 5% dextrose). Carbon dioxide has the same refractive index as air, provides optimal visibility (Van der Pas and Vancaille, 1990Go) and, as a gas, is easy to infuse, although vision can be impaired by bubbles or bleeding during the procedure.

Few objective reports have been published (Nagele et al., 1996aGo; Pellicano et al., 2003Go) on the use of liquids for outpatient hysteroscopy, particularly low-viscosity fluids such as normal saline (0.9% sodium chloride). The attraction of these solutions is their ready availability, insufflation by use of a simple pressure bag, rapid reabsorption from the peritoneal cavity in case of transtubal leakage, good vision, low viscosity, and miscibility in the blood (Nagele et al., 1996aGo).

The aim of this prospective, randomized study was to measure patients’ discomfort using a visual analogue scale (VAS) pain score after hysteroscopy with either CO2 or normal saline.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Between January and May 2002, diagnostic hysteroscopy was performed in 415 patients who were randomized into two groups according to the distension medium used (CO2, n = 201; normal saline, n = 214). The nature of randomization was to alternate the distension medium between CO2 and normal saline on a weekly basis.

Indications for hysteroscopy included the following: abnormal uterine bleeding (AUB), thick endometrium or suspected intracavitary pathology such as endometrial polyps or myomas at ultrasonography, infertility, tamoxifen therapy and cytological endometrial hyperplasia.

Hysteroscopy was performed using a 2.9 mm-diameter hysteroscope with a 30° foreoblique lens. An electronic hysteroflator of CO2 (Storz, Tutlingen-Germany) was used with a pressure not exceeding 100 mmHg and a flow rate of 40 ml/min. Normal saline was infused with a 100 mmHg pressure bag. All procedures were performed by the same operator, and no local anaesthesia or systemic drugs were given to any patient.

The cervix was visualized using a speculum, after which the hysteroscope was slowly introduced into the cervical canal in order to evaluate carefully the endometrial surface and to visualize the tubal ostia. The endocervical canal was inspected during extraction of the hysteroscope. The same procedure was performed in the CO2 and normal saline groups. When indicated, endometrial biopsy was performed using a Novak curette. The procedure time was measured from insertion of the hysteroscope into the external uterine orifice until its removal.

At ~10 min after hysteroscopy, women were asked by another operator, who was not directly involved with the procedure, to rate the pelvic pain experienced on a 100 mm VAS (0 = no pain; 100 = worst imaginable pain). The results of pain scores were expressed as mean ± SD, where a score of 0–40 indicated minimal pain, 41–70 indicated moderate pain, and 71–100 indicated severe pain.

Overall pelvic discomfort after the procedure was also assessed. Pelvic discomfort observed during the hysteroscopic procedure was mainly due to passage of the hysteroscope through the internal cervical os and to the induced uterine contraction that is determined by uterine cavity distension.

Satisfaction rate was evaluated on a 5-point scale, where 0 = no satisfaction, 1 = mild satisfaction, 2 = moderate satisfaction, 3 = satisfaction, and 4 = maximum satisfaction. Women were also interviewed about the presence or absence of shoulder tip pain.

The data were analysed using Student’s t-test; a P-value < 0.05 was considered to be statistically significant.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
The characteristics of the patients are listed in Table I. The mean age of women in the CO2 group was 50.1 ± 12.4 (range 20–78) years, while that for women in the normal saline group was 48.2 ± 13.1 (range 24–87) years.


View this table:
[in this window]
[in a new window]
 
Table I. Main characteristics of patients
 
AUB was the most frequent indication in both the CO2 and normal saline groups (44.3 versus 43%), followed by suspected intracavitary pathology (20.8 versus 20.9%), thick endometrium (11.9 versus 13.5%), infertility (7.5 versus 11.7%), tamoxifen therapy (9 versus 7.2%) and cytological endometrial hyperplasia (6.5 versus 3.7%) (Table II).


View this table:
[in this window]
[in a new window]
 
Table II. Indications to hysteroscopy
 
The findings of intrauterine pathologies were similar in the two groups. A normal endometrium was observed in 69 (34.3%) of the CO2 group compared with 66 (30.8%) of the saline group. Comparative respective values were 56 (27.9%) versus 62 (29%) for polyps, 29 (14.4%) versus 30 (14%) for myomas, and 15 (7.5%) versus 25 (11.7%) for Mullerian anomalies. One case of endometrial cancer was detected in each group (0.5% in CO2 versus 0.5% in normal saline) (Table III).


View this table:
[in this window]
[in a new window]
 
Table III. Hysteroscopic findings
 
Endometrial biopsy was performed in 33 (16.4%) of the 201 patients with CO2 and in 34 (15.9%) of the 214 with normal saline (P = NS). A successful outcome was achieved in all 415 hysteroscopies, and none of the women converted from CO2 to normal saline.

The mean procedure time was 115 ± 12 s in the CO2 group compared with 90 ± 10 s in the normal saline group (P < 0.01).

Irrespective of the distension medium used, pelvic discomfort was worse in nulliparous women (score 39.0 ± 26.5) than in multiparous women (score 30.4 ± 25.9) (P < 0.05; combined pre- and postmenopausal data not shown in table), especially if they were premenopausal (Table IV). However, when taking into consideration all patients and both distension media, pelvic discomfort was worse with normal saline than with CO2 (score 36.2 ± 26.8 versus 29.4 ± 25.4, P < 0.05; data not shown), especially if the women were nulliparous or premenopausal (Table V).


View this table:
[in this window]
[in a new window]
 
Table IV. Correlation of parity, menopausal status and pelvic discomfort
 

View this table:
[in this window]
[in a new window]
 
Table V. Correlation between distension medium and pelvic discomfort
 
Pelvic discomfort was seen to be worse in the normal saline group than in the CO2 group (42.3 ± 26.6 versus 31.3 ± 24.4, P < 0.05), but only if the women were premenopausal nulliparous (Table VI). By correlating parity and distension medium, pelvic discomfort was seen to be worse in nulliparous than multiparous women (43.6 ± 27.6 versus 32.6 ± 25.8, P < 0.05), but only among the normal saline group (data not shown).


View this table:
[in this window]
[in a new window]
 
Table VI. Correlation between parity, menopausal status and distension medium
 
The satisfaction rate was significantly higher in the normal saline group (3.4 ± 0.6) compared with the CO2 group (2.2 ± 0.7) (P < 0.05). Likewise, due to the introduction of a 2.9 mm hysteroscope, there were no reports of any vasovagal reaction in either group. Neither tenaculum nor cervical dilatation were used in either group, and shoulder tip pain was reported by only in 3.9% of patients in the CO2 group.


    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Hysteroscopy is certainly the most accurate diagnostic procedure to monitor intrauterine lesions, with preference for the procedure demonstrated by the increasing numbers of requests to study women presenting with AUB and infertility, as well as those undergoing hormone replacement and tamoxifen therapy (Valle, 1980Go; Rafael and Valle, 1999Go).

Improvements in technology resulting in the development of narrow-diameter hysteroscopes have made it possible to perform hysteroscopy both feasibly and safely on an outpatient basis (Nagele et al., 1996bGo; Wieser et al., 1998Go).

Nowadays, CO2 is the most commonly used distension medium, mainly on the basis of its properties of excellent endoscopic visualization of the uterine cavity and low risk of side effects (Loffer, 1993Go), although the formation of bubbles and/or the presence of blood sometimes leads to an inadequate diagnostic examination (Pellicer and Diamond, 1988Go). Some authors use low-viscosity fluids based on low cost, an absence of any side effects, simplicity of use and good vision. The advantages of this approach are a good visualization of the uterine cavity in the presence of blood clots, mucus and debris (Nagele et al., 1996aGo). In cases of AUB [one of the most common indications for hysteroscopy (40%), and ~44% in the present series], the use of low-viscosity fluids may be advantageous when exploring the uterine cavity (Valle, 1981Go; Mencaglia et al., 1987Go; Soderstrom, 1992Go; Perez-Medina et al., 2000Go).

When comparing the two most used distension media, one group (Nagele et al. 1996aGo) reported that the hysteroscopic procedure time with normal saline was generally shorter than with gas. The present data confirm that normal saline is certainly an excellent distension medium with a short procedure time (90 ± 10 s). Moreover, it allows good visualization and a more rapid distension compared with CO2, which requires a longer time to expand the uterine cavity, albeit in a characteristically different manner.

With regard to the patient’s discomfort during hysteroscopy, both distension media cause only minimal to moderate pelvic pain, and this permits the operator to perform the procedure with good patient compliance (Mantha et al., 1993Go; Giorda et al., 2000Go).

In the present series, normal saline as a distension medium evoked a more intense pelvic discomfort than did CO2 (pain scores 36.2 ± 26.8 versus 29.4 ± 25.4; data not shown), especially in nulliparous or premenopausal patients. By contrast, another group recently reported (albeit in a smaller sample size than the present series; n = 189) a significant reduction in abdominal pain after distension with normal saline compared with CO2 (Pellicano et al., 2003Go). However, the same authors did not find any statistically significant difference between the two groups in terms of postoperative pain at 2 h after hysteroscopy. At present, both distension media provide good quality of vision, handling and safety, and are comparable in terms of cost and patient discomfort, which was shown to be minimal for both groups in the present series.

Unfortunately, our study did not use an adequate method for patient randomization. In fact, the method of randomization, which was to alternate the type of distension medium on a weekly basis, does not follow CONSORT (Consolidated Standards of Reporting of Trials) guidelines (Schultz et al., 1994Go; Grimes, 2002Go).

In recent years, the use of hysteroscopy has been extended widely on the basis of its diagnostic accuracy and ease of performance. These aspects constitute important benefits for women who undergo this outpatient procedure, for which compliance is high and hospital costs are reduced (Valle, 1988Go).

In conclusion, when considering the high frequency of AUB as an indication for hysteroscopic examination and the significantly shorter procedure time, normal saline should be considered the most appropriate distension medium for outpatient hysteroscopy.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Baggish, M.S. (1989) Distending media for panoramic hysteroscopy. In Baggish, M.S., Barbot, J. and Valle, R.F. (eds), Diagnostic and Operative Hysteroscopy. Year Book Medical Publishers, Inc., Chicago, pp. 89–93.

DeIaco, P., Marabini, A., Stefanetti, M., Del Vecchio, C. and Bovicelli, L. (2000) Acceptability and pain of outpatient hysteroscopy. J. Am. Assoc. Gynecol. Laparosc., 7, 71–75.[ISI][Medline]

Giorda, G., Scarabelli, C., Franceschi, S. and Campagnutta, E. (2000) Feasibility and pain control in outpatient hysteroscopy in postmenopausal women: a randomized trial. Acta Obstet. Gynecol. Scand., 79, 593–597.[CrossRef][ISI][Medline]

Grimes, D.A. (2002) The ‘CONSORT’ guidelines for randomized controlled trials in obstetrics and gynecology. Obstet. Gynecol., 100, 631–632.[Free Full Text]

Loffer, F.D. (1993) Complications from uterine distension during hysteroscopy. In Corfman, R.S., Diamond, M.P. and DeCherney, A. (eds), Complications of Laparoscopy and Hysteroscopy. Blackwell Scientific Publications, Boston, pp. 176–185.

Mantha, S., Thisted, R., Foss, J., Ellis, J.E. and Roizen, M.F. (1993) A proposal to use confidence intervals for visual analog scale data for pain measurement to determine clinical significance. Anesth. Analg., 77, 1041–1047.[Abstract]

Mencaglia, L., Perino, A. and Hamou, J. (1987) Hysteroscopy in perimenopausal and postmenopausal women with abnormal uterine bleeding. J. Reprod. Med., 32, 577–582.[ISI][Medline]

Nagele, F., Bournas, N., O’Connor, H., Broadbent, M., Richardson, R. and Magos, A. (1996a) Comparison of carbon dioxide and normal saline for uterine distension in outpatient hysteroscopy. Fertil. Steril., 65, 305–309.[ISI][Medline]

Nagele, F., O’Connor, H., Davies, A., Badawy, A., Mohamed, H. and Magos, A. (1996b) 2500 outpatient diagnostic hysteroscopies. Obstet. Gynecol., 88, 87–92.[Abstract/Free Full Text]

Pellicano, M., Guida, M., Zullo, F., Lavitola, G., Cirillo, D. and Nappi, C. (2003) Carbon dioxide versus normal saline as a uterine distension medium for diagnostic vaginoscopic hysteroscopy in infertile patients: a prospective, randomized, multicenter study. Fertil. Steril., 79, 418–421.[CrossRef][ISI][Medline]

Pellicer, A. and Diamond, M.P. (1988) Distending media for hysteroscopy. Obstet. Gynecol. Clin. North Am., 15, 23–28.[ISI][Medline]

Perez-Medina, T., Bajo, J.M., Martinez-Cortes, P., Castellanos, P. and Perez de Avila, I. (2000) Six thousand office diagnostic-operative hysteroscopies. Int. J. Gynaecol. Obstet., 71, 33–38.[CrossRef][ISI][Medline]

Rafael, F. and Valle, R.F. (1999) Office hysteroscopy. Clin. Obstet. Gynecol., 42, 276–289.[CrossRef][ISI][Medline]

Schultz, K.F., Chalmers, I., Grimes, D.A. and Altman, D.G. (1994) Assessing the quality of randomization from reports of controlled trials published in obstetrics and gynecology journals. JAMA, 272, 125–128.[Abstract]

Soderstrom, R.M. (1992) Distending the uterus: what medium is best? Clin. Obstet. Gynecol., 35, 225–228.[ISI][Medline]

Valle, R.F. (1980) Hysteroscopy in the evaluation of female infertility. Am. J. Obstet. Gynecol., 137, 425–431.[ISI][Medline]

Valle, R.F. (1981) Hysteroscopic evaluation of patients with abnormal uterine bleeding. Surg. Gynecol. Obstet., 153, 521–526.[ISI][Medline]

Valle, R.F. (1988) Future growth and development of hysteroscopy. Obstet. Gynecol. Clin. North Am., 15, 111–126.[ISI][Medline]

Van der Pas, H. and Vancaille, T. (eds) (1990) Gynecologic Endoscopy. Volume 2. Manual of Hysteroscopy. Elsevier Science Publishers, Amsterdam, pp. 1–7.

Vercellini, P., Colombo, A., Mauro, F., Oldani, S., Bramante, T. and Crosignani, P.G. (1994) Paracervical anesthesia for outpatient hysteroscopy. Fertil. Steril., 62, 1083–1085.[ISI][Medline]

Wieser, F., Kurz, C., Wenzl, R., Albrecht, A., Huber, J.C. and Nagele, F. (1998) Atraumatic cervical passage at outpatient hysteroscopy. Fertil. Steril., 69, 549–551.[CrossRef][ISI][Medline]

Submitted on January 10, 2003; resubmitted on May 16, 2003; accepted on August 1, 2003.