Introduction

In 1869, Pantaleoni performed the first documented hysteroscopic procedure when he used his “endoscope” for visualization of the uterine cavity, and he treated an endometrial polyp in a 60-year-old woman with postmenopausal bleeding . It is notable that even this first case report of hysteroscopy commented on the risk of hysteroscopic complications and how to avoid them, “…she was placed as near as possible to the borders of the bed so that Dr Cruise’s light source did not set fire to the curtains…”, which illustrates the subsequent technological advances made in endoscopic illumination.

Although the surgical dictum “Primum non nocere” should be respected, every operation has risks of complications. These may be divided into general risks, for example, anesthesia problems, and risks specific to the way the operation is performed. Minimal-access surgery (MAS) is no exception; hysteroscopic procedures involve introduction of instruments into the uterus and distension with media in a fashion not used conventionally. As a result, hysteroscopic surgery has specific iatrogenic risks, particularly during the learning phase of the surgeon.

This review focuses on practical and technological advances whereby complications may be decreased during hysteroscopic surgery.

The scale of the problem

Obstetrics and gynecology has a long tradition of assessing and evaluating patient outcomes; MAS was born into this tradition and the result is a large volume of data in the scientific literature that enables us to benefit from a vast range of clinical experience. The formation of national supervisory bodies, for example, the UK Royal College of Obstetricians (RCOG), the British Society for Gynaecological Endoscopy (BSGE), the American Association of Gynecologic Laparoscopists (AAGL), and the Australasian Gynaecological Endoscopy & Surgery Societies (AGES), provided the impetus for structured training and accreditation. As new hysteroscopic surgical techniques have developed, these agencies have played a leading role in ensuring their safe and appropriate use (e.g., MISTLETOE) . Increasingly, countries use large data series to assess the safety and efficacy of new surgical technologies. These national databases are listed in Table 1 .

One of the weaknesses of the current dataset reporting systems is that there are no denominator data on which to determine the incidence, risk, or adverse outcome. An example of this relates to the still unanswered question of which second-generation global endometrial ablation device is the safest – all of these devices have had adverse outcomes, but there are no definitive data on which is the safest .

What are the established rates of complications associated with hysteroscopic surgery? Different procedures have different complication rates. There have been a number of large-scale national audits performed ( Table 2 ) . Jansen et al. published a Dutch national audit in 2000, which examined >11,000 diagnostic and 2500 operative hysteroscopies . The audit found that there were significantly more complications during operative hysteroscopy compared to diagnostic. Clinically significant fluid overload occurred in 0.2% of operative hysteroscopies – particularly during myomectomy resections. Although uterine perforations occurred during both diagnostic (incidence 0.13%) and operative hysteroscopy (incidence 0.76%), there were no statistically significant differences between entry- and technique-related causes of uterine perforation. Perforations occurred most commonly during dilatation (70%) for diagnostic hysteroscopy. The riskiest operative hysteroscopic procedure was adhesiolysis (risks of complication 4.5%), compared to polypectomy (0.4%). Jansen’s paper reported no deaths due to hysteroscopic surgery during their study .

Minimal-access procedure-related complications are relatively infrequent. However, there is no room for complacency; rare but very serious complications for commonly performed procedures such as endometrial ablation can still represent a major problem on an international scale.

For this article, we have examined the evidence base for the following aspects of performing safe hysteroscopic surgery:

• •
  

  Appropriate case selection and recognition of the learning phase.

  
  
• •
  

  Patient preparation.

  
  
• •
  

  Distension media and hysteroscopic equipment.

The scale of the problem

Obstetrics and gynecology has a long tradition of assessing and evaluating patient outcomes; MAS was born into this tradition and the result is a large volume of data in the scientific literature that enables us to benefit from a vast range of clinical experience. The formation of national supervisory bodies, for example, the UK Royal College of Obstetricians (RCOG), the British Society for Gynaecological Endoscopy (BSGE), the American Association of Gynecologic Laparoscopists (AAGL), and the Australasian Gynaecological Endoscopy & Surgery Societies (AGES), provided the impetus for structured training and accreditation. As new hysteroscopic surgical techniques have developed, these agencies have played a leading role in ensuring their safe and appropriate use (e.g., MISTLETOE) . Increasingly, countries use large data series to assess the safety and efficacy of new surgical technologies. These national databases are listed in Table 1 .

One of the weaknesses of the current dataset reporting systems is that there are no denominator data on which to determine the incidence, risk, or adverse outcome. An example of this relates to the still unanswered question of which second-generation global endometrial ablation device is the safest – all of these devices have had adverse outcomes, but there are no definitive data on which is the safest .

What are the established rates of complications associated with hysteroscopic surgery? Different procedures have different complication rates. There have been a number of large-scale national audits performed ( Table 2 ) . Jansen et al. published a Dutch national audit in 2000, which examined >11,000 diagnostic and 2500 operative hysteroscopies . The audit found that there were significantly more complications during operative hysteroscopy compared to diagnostic. Clinically significant fluid overload occurred in 0.2% of operative hysteroscopies – particularly during myomectomy resections. Although uterine perforations occurred during both diagnostic (incidence 0.13%) and operative hysteroscopy (incidence 0.76%), there were no statistically significant differences between entry- and technique-related causes of uterine perforation. Perforations occurred most commonly during dilatation (70%) for diagnostic hysteroscopy. The riskiest operative hysteroscopic procedure was adhesiolysis (risks of complication 4.5%), compared to polypectomy (0.4%). Jansen’s paper reported no deaths due to hysteroscopic surgery during their study .

Minimal-access procedure-related complications are relatively infrequent. However, there is no room for complacency; rare but very serious complications for commonly performed procedures such as endometrial ablation can still represent a major problem on an international scale.

For this article, we have examined the evidence base for the following aspects of performing safe hysteroscopic surgery:

• •
  

  Appropriate case selection and recognition of the learning phase.

  
  
• •
  

  Patient preparation.

  
  
• •
  

  Distension media and hysteroscopic equipment.

Case selection and the learning phase

We should acknowledge that not all cases are suitable for hysteroscopic surgery; what may be technically possible may not necessarily be the safest option, for example, hysteroscopic surgery for large (>5-cm) type II submucous fibroids ( Table 3 ). Whilst it is feasible for very experienced gynecological endoscopists to perform these procedures hysteroscopically, it is important to appreciate that what may be achieved by a few gifted surgeons may not be readily accomplished by the average practicing, general gynecologist.

Proper case selection is of particular importance in the “learning phase” – as a rule, start with simple cases. When performing hysteroscopic resection of fibroids, start with small type 0 or type I submucous fibroids ( Table 3 ). Once experience and confidence have been gained, more difficult cases may be undertaken.

In the Dutch audit by Jansen et al., there were no obvious risks based on the patient’s menopausal status, but this may have been affected by many of the premenopausal women undergoing operative hysteroscopy undergoing pre-treatment with gonadotropin-releasing hormone analogues (GnRH-a) .

Surgical experience is an obvious factor to consider, but this is subject to considerable inherent bias as the most difficult procedures tend to be performed by those with the most surgical experience. Jansen’s study demonstrated that, for diagnostic hysteroscopies, there were no significant differences in the number of complications and surgical experience. However, for operative procedures, most complications occurred when the surgeons had more operative experience – 37% of all operative hysteroscopic complications occurred in the group in which surgeons had performed >50 of the specific procedures . This somewhat counterintuitive finding underlines the importance of continuous audit of surgical practice, and the appropriate emphasis on surgical training and accreditation.

The UK RCOG has classified hysteroscopic procedures in terms of surgical complexity (and hence risk) in order to guide accreditation and training in hysteroscopic surgery ( Table 4 ) .

Patient preparation

Appropriate patient preparation involves more than positioning the patient when performing the surgery. Patient preparation should start from the patient’s first presentation; one of the most important factors in patient satisfaction and avoiding litigation is a fully informed patient, particularly if there are known risk factors (e.g., adhesions) to performing surgery .

There is a role for hysteroscopic surgery in treating endometrial polyps and fibroids as a means to improve fertility . However, patients may have unrealistic expectations about their surgery; a leaflet explaining the procedure is a useful adjunct to a full discussion. A plan of the “worst-case scenario” should be made, for instance, the need for a laparotomy or hysterectomy in the event of a uterine perforation. When complications do occur, this should be fully documented, and remedial surgery instituted as soon as possible .

For operative hysteroscopy, there is much discussion in the literature on endometrial preparation prior to endometrial ablation. However, there is no consensus on the optimal method of endometrial preparation (GnRH-a, Danazol, or the combined oral contraceptive) and whether these agents have any benefits for second-generation endometrial ablation, or other hysteroscopic surgery such as removal of septae or submucous fibroids . GnRH analogues are very potent drugs, producing menopausal symptom side effects. Another consequence of their use is that they may cause difficulties during cervical dilatation .

There have been some studies assessing the use of cervical “ripening” agents prior to hysteroscopy; these are typically the prostaglandin analogue misoprostol, the progesterone antagonist mifepristone, or osmotic dilators (laminaria stents). The published literature demonstrates no benefit in using cervical preparation for those patients undergoing diagnostic hysteroscopy . For those patients undergoing operative hysteroscopy and cervical dilatation beyond 5-mm diameter, these agents may be beneficial especially in premenopausal women .

Another aspect of patient preparation is analgesia. There are obvious differences in the analgesic requirements depending on the nature and duration of the hysteroscopic procedure. If we assess the evidence base for outpatient-based hysteroscopic procedures, the systematic review and meta-analysis by Cooper et al. in 2010 established that, if local anesthetic is needed, paracervical injection is the best method of pain control .

The question of whether antibiotic prophylaxis minimizes infectious morbidity during hysteroscopic procedures has been the subject of a Cochrane review . The authors found no robust studies investigating the effect of antibiotic prophylaxis on infectious complications. A pseudo-randomized French study examined the role of antibiotic prophylaxis for diagnostic hysteroscopy and found no benefits. This concurs with the multicenter study by Aydeniz et al. , which quoted a very low rate of endomyometritis (0.01%) after operative hysteroscopic surgery in a case series of >21,000 procedures, although the authors noted that some of the units reporting the data used antibiotic prophylaxis routinely.