Discussion
Decisions concerning any procedure under general anesthesia are less easily made due to the associated comorbidities and higher risks among women older than 70 years of age. The performance of a procedure in the OR may also be affected by the unwillingness of the patient to undergo general anesthesia when given the option, even at the cost of not reaching a diagnosis.
Altogether, 194 (87.4%) patients in our cohort who were referred to office hysteroscopy had a successful procedure. Our results showed a success rate similar to the 74.5% success rate for office hysteroscopy in postmenopausal women recently reported by Sauvan et al4. Sousa et al., who used a hysteroscopic technique similar to ours on 88 postmenopausal patients with genital bleeding, reported similar failure rates of 17.8%5.
Endometrial sampling was generally less likely to be performed in the office compared to the OR, but the fact that more than one-half of the women in our cohort (58.6%) did have a biopsy taken during the office procedure is of considerable diagnostic value for this age group. It is important to emphasize that even incomplete procedures, which were much more common in the office group than in the OR group, are of some value in this age group, especially if a biopsy has been carried out. For example, if a biopsy from a benign endometrial polyp is taken, the patient and her physician would feel more confident about choosing a conservative sonographic follow-up, thus avoiding a procedure under anesthesia. With good medical practice becoming more and more tailored to the individual needs of the patient, we should be aware of this option for minimizing risks to the patient.
Bleeding and large or multiple findings were a leading cause of failed or incomplete procedures in the office, while in the OR they limited the procedure in only two cases. Bleeding in the uterine cavity limits the operating viewing field during hysteroscopy, but this problem is overcome in the OR by washing the uterine cavity continuously with an irrigation flow pump using higher pressures than those used in the office. This also serves to explain the significant difference in the indications for which women were referred to the treatment options. More women were referred to office procedures due to abnormal sonographic findings, while more women with active PMB were referred to procedures in the OR.
Our results revealed that cervical stenosis is not a contraindication for office hysteroscopy. We managed to overcome cervical stenosis in more than half of the patients who presented with it (32/54 cases, 59%). The techniques used in our clinic to contend with cervical stenosis are either mechanical, including forcing pressure with the hysteroscope or using scissors, or electrical by means of a bipolar electrode for widening the orifice and penetrating through the fibrotic tissue. Pretreatment with 25 mcg of vaginal estradiol in combination with 400 mcg of vaginal misoprostol before the procedure can facilitate passage of the hysteroscope through the cervical canal6.
Malignancy was found in 30.9% of the patients who presented with PMB, as well as in 7.3% of the patients who had no PMB. The pathological results from the repeat hysteroscopy are striking. One-quarter (12/48) of the samples retrieved from repeat hysteroscopy were abnormal. This emphasizes the importance of pursuing a diagnosis. In selected cases of a moderate-to-high index of suspicion, we do not favor a conservative sonographic follow-up but rather recommend referring the patient directly to a repeat hysteroscopy. It is noteworthy that malignancy was detected in 6/36 (16.7%) of the patients who did not present with PMB and who were referred to a repeat hysteroscopy.
The volume of procedures in the office increased 16 times during the study period, and a higher percentage of malignancies was diagnosed compared to office procedures (Figure 1B ). These trends are the result of a newly implemented technique in our institution, and also of a change of clinical practice of referring women to the office option as a first-line procedure. Once the refined biopsy-taking technique was implemented, no significant change in yield was observed during the last three years of the study, indicating that the technique is relatively quickly and easily learned and put into practice.
The most compelling advantage of office hysteroscopy compared to OR hysteroscopy in this cohort is the absence of complications in the former and the rate and severity of complications in the latter, including ICU hospitalization, sepsis, abdominal operation, and two mortalities related to the OR hysteroscopy procedure. A similar recent publication on office hysteroscopy in postmenopausal women also did not observe any complications, such as uterine infection, uterine perforation, or genital tract hemorrhage4.
Our study has several limitations. The main limitation is that the two sub-groups were not matched for demographic and clinical characteristics or for indications for undergoing the procedure. Also, we did not have information on previous procedures that may have been performed outside our medical center. It is likely that some patients with a history of failure of a previous trial of office procedures were directly referred to OR hysteroscopy. While our data are primarily descriptive, they nevertheless suggest that office hysteroscopy for women over the age of 70 is safe, feasible, and reliable. Moreover, to the best of our knowledge, this is the first study to compare office and OR hysteroscopy for this age group. We believe that patient selection will improve with greater application of office hysteroscopy and as more and more physicians become aware of its advantages as well as its limitations. This, in turn, will lead to higher rates of successfully completed procedures and to higher rates of accurate pathological diagnoses in the office setting. Further prospective trials are needed in order to confirm our findings. Such studies may help to design a tool for better patient selection for either office or OR hysteroscopy.