Diagnostic hysteroscopy is a commonly performed gynecologic procedure to evaluate the endometrial cavity. This article focuses on the procedure of diagnostic hysteroscopy. Other excellent Medscape Reference articles include an overview of hysteroscopy by Drs. Petrozza and Attaman and operative hysteroscopy by Drs. Crochet, Yeh, and Price.
Broadly, 2 systems of diagnostic hysteroscopy exist: panoramic (also known as direct optical) and contact (also known as contact microhysteroscopy). Modern references to hysteroscopy usually imply a panoramic technique in which the uterine cavity is distended with liquid or gas and evaluated with the hysteroscope. Contact hysteroscopy is a related procedure in which no distending media is used and the hysteroscope is passed directly into the uterus and put in gentle contact with the endometrial lining to obtain maximum magnification. Only tissue that is in direct contact with the distal tip can be inspected and evaluated. External light sources can be used but some descriptions of the procedure rely only on ambient light.  This technique is much slower, and the blunt trauma to the uterine cavity can lead to higher rates of uterine perforations.
An image depicting diagnostic hysteroscopy can be seen below.
Diagnostic hysteroscopy is used to evaluate the endocervical canal, endometrial cavity, and tubal ostia. The procedure is often coupled with sight-directed biopsy or followed by endometrial curettage to evaluate for endometrial pathology.
Other methods of evaluating the female reproductive tract include pelvic ultrasonography with or without saline infusion into the endometrial cavity, hysterosalpingography (HSG), and endometrial sampling. Unlike these procedures, hysteroscopy has the unique advantage of combining a thorough diagnostic procedure with treatment.
The following conditions can be evaluated with diagnostic hysteroscopy and further treated with operative hysteroscopic techniques. Indications for diagnostic hysteroscopy include:
Abnormal premenopausal or postmenopausal uterine bleeding
Removal of foreign body (intrauterine device, retained fetal bone) 
Confirmation of abnormal test findings (abnormal HSG or thickened endometrial lining on sonography)
Suspected Müllerian anomalies
Few contraindications to hysteroscopy exist. Severe medical comorbidities may be a contraindication and should be assessed for each individual’s needs and tolerances. Contraindications to hysteroscopy include the following:
Viable intrauterine pregnancy
Female genital tract cancer, including cervical or uterine cancer
Active pelvic infection
Pregnancy in the premenopausal patient should always be excluded with serum or urine β-human chorionic gonadotropin (hCG) testing prior to surgery. Hysteroscopy should not be performed in the setting of cervical or endometrial cancer. A recent meta-analysis supports the long-held belief that cancer cells are disseminated into the peritoneal cavity by the intrauterine pressure of hysteroscopy.  Lastly, treatment of pelvic infections is mandatory prior to hysteroscopy and can decrease the chance of propagating gynecologic infections that can lead to pelvic pain, tubal factor infertility, and even death. [4, 5]
Severe pain and patient anxiety are among the most common causes of surgical failure.  Many anesthetic options are available to patients undergoing hysteroscopy. A Cochrane review supports the use of local anesthesia as effective pain control during and within 30 minutes of completing hysteroscopy. The same review did not show significant pain reduction with the widely practiced use of NSAIDs or opioids during or after the procedure. 
Conventional panoramic hysteroscopy requires some form of anesthesia, while the smaller caliber flexible hysteroscopes require little to no anesthesia. For hysteroscopes of larger diameter, injectable local anesthetics combined with preoperative vaginal misoprostol (Cytotec) is usually sufficient. Occasionally, regional anesthesia, monitored anesthesia care (MAC), or general anesthesia may be indicated for more extensive procedures or for patients who have lower pain tolerance and/or anxiety.
Many practitioners favor the use of topical anesthesia, although studies have shown mixed efficacy. Aerosolized preparations of lidocaine may decrease cervical pain from tenaculum placement but do not decrease uterine sensation.  Additionally, a study comparing the addition of lidocaine with the saline distension media showed no difference in pain score compared with saline alone.  In contrast, transcervical instillation of 5 mL of 2% mepivacaine lowered pain scores and decreased the rate of vasovagal reactions for women undergoing diagnostic hysteroscopy followed by endometrial biopsy.  Topical anesthetics typically do not provide long-lasting relief but may be sufficient for the nonanesthetized patient. [8, 11, 12, 13]
Infiltration of the paracervical tissue with a local anesthetic is commonly used for hysteroscopic anesthesia. A paracervical block can decrease the pain of tenaculum placement, cervical dilation, and hysteroscope insertion through the cervix. However, paracervical anesthesia has less effect on the pain of uterine distension.  One must balance the expected pain of the hysteroscopic procedure with the pain and potential side-effects of the paracervical block, which include bradycardia and hypotension.  For these reasons, many providers choose to forgo this step, especially for brief diagnostic procedures. Common anesthetic agents are 1% lidocaine, mepivacaine, prilocaine, ropivacaine, bupivacaine, and etidocaine. Of these, bupivacaine and etidocaine have longer durations and can last upwards of 2-3 hours. [16, 17]
A recent study suggests that paracervical ropivacaine controls intraoperative pain slightly better than lidocaine during surgical abortions.  In most cases, 10 mL of bupivacaine 0.25%, mepivacaine 1%, or lidocaine 1-2% is an adequate volume for paracervical anesthesia. If more volume is required, understanding the safe doses for each particular agent is important. For example, the toxic dose of lidocaine is approximately 4.5 mg/kg. [15, 16, 19] For a 70-kg woman, 4.5 mg/kg x 70 kg = 315 mg, or 31.5 mL of lidocaine 1%.
Nonsteroidal anti-inflammatory drugs
Despite the lack of strong evidence to encourage the use of NSAIDs, the use of oral or intravenous analgesics in the preoperative period is common. Smaller studies of NSAID use have been shown to reduce only postoperative pain with minimal effect on intraoperative pain and still may be a good choice for select patients. [20, 21] Tramadol IV is one option in the intraoperative and postoperative period for women undergoing diagnostic hysteroscopy and endometrial biopsy without cervical dilation. 
Patients with cervical stenosis may require some cervical dilation for diagnostic hysteroscopy, especially after menopause. The administration of analgesia does not decrease the number of aborted procedures due to cervical stenosis.  Misoprostol, a synthetic prostaglandin E1 analogue, may be administered orally or vaginally for cervical ripening to improve the likelihood of successful cervical dilation and decrease intraoperative pain with few adverse events. [23, 24] Randomized trials on the use of misoprostol have shown conflicting results and suggest that misoprostol is perhaps less effective in postmenopausal women. 
The basic equipment used to perform diagnostic hysteroscopy includes the hysteroscope, light source, and distention media. Modern systems use a digital camera that attaches to the eyepiece and transmits images to a visual monitoring screen.
Diagnostic hysteroscopes are available in different sizes: the choice of a particular type can affect pain tolerance, image quality, and intraoperative surgical options. The hysteroscope usually includes the telescope and the sheath that encases it. Newer diagnostic models, however, may lack the sheath. Outer sheaths have accessory channels that enable the inflow and outflow of the distention media. When applicable, the sheath and a separate connecting bridge also have ports to insert operative and manipulating instruments.
The 2 instrument dimensions of great consequence are the outer diameter (OD) and working length (WL).
The outer diameter indicates the width of the barrel including the sheath. For diagnostic hysteroscopes that do not use a sheath, the OD refers only to the width of the optic barrel. Modern hysteroscopes range from 2.7-10 mm; the OD of a diagnostic scope typically ranges from 2.7-5.0 mm, while the OD of an operative sheath ranges from 5-10 mm. Rigid scopes with an OD exceeding 5.0 mm usually require some degree of cervical dilation, while use of a narrow caliber (or flexible) hysteroscope usually requires no cervical dilation. Not surprisingly, the OD also correlates with the amount of operative pain. In a study of hysteroscopy in postmenopausal women, subjects undergoing small diameter hysteroscopy (<3.5 mm) without local injection had less pain than those with paracervical blocks undergoing the procedure with a 5.0 mm hysteroscope. 
The working length measures the distance between the distal lens to the proximal eyepiece and ranges from 160-302 mm.  Longer working lengths permit the surgeon to operate farther from the vagina and may be of strategic importance in the morbidly obese patient. Working lengths typically do not vary enough to produce differences in picture quality between hysteroscopes of similar build and quality.
The 3 parts of the hysteroscope (excluding the sheath) are the lens, the barrel, and the eyepiece. The depth of field on the scope is usually between 2-3 cm, and magnification can reach up to 35X with certain liquid distention media and lens positioning.  Hysteroscopes are available in a variety of viewing angles: 0°, 12°, 15°, 25°, 30°, and 70°. Zero degree scopes provide a wide-angled view in-line with the barrel. See the image below.
Angled scopes allow for clear views of the periphery without requiring excessive operator movement and are often helpful to visualize the tubal ostia. Cameras that attach to the eyepiece are frequently used to transmit images to a video monitor. Electronic displays allow the operating room personnel to watch the procedure and are convenient for intraoperative teaching. These systems are also capable of electronically documenting the procedure with still images and video clips for future reference.
Rigid versus flexible
Hysteroscopes can be classified into rigid and flexible (or semirigid) types. Rigid hysteroscopes require some assembly. Most designs require that the telescope be inserted into a sheath that is then attached to a bridge. Bridges for a diagnostic scope may only have a single inflow port. Operative bridges typically have 2 media ports with additional ports for instruments. Dual media ports for inflow and outflow can produce a steady laminar flow to improve image clarity during procedures in which blood can obscure the field of view. See the images below.
A growing number of very narrow, flexible hysteroscopes that produce less pain and are more suitable for outpatient applications are available. Flexible scopes are available in diagnostic and operative models and can have an OD as small as 2.7 mm. In a recent trial randomizing 144 women to either flexible or rigid diagnostic hysteroscopy, women in the rigid group experienced significantly more pain than those in the flexible group. Physicians, however, reported better optical quality with the rigid scope when compared with the flexible scope. 
See the image below.
A light source is required to illuminate the intrauterine cavity. The intensity of light is generally lower than what is used in laparoscopy because hysteroscopic viewing distances are less than 5 cm. Reusable fiberoptic cables are the most common method of light transmission. Fiberoptic cables provide lossless light transmission without generating excessive heat. Lighting sources include tungsten, metal halide, and xenon. A xenon lighting system attached to a liquid cable is considered the superior option. 
Distention media and management
The intrauterine cavity is a potential space; hysteroscopic examination requires the cavity be distended with either gas or liquid. The use of hysteroscopic monopolar or bipolar energy limits the choice of distention media. Because diagnostic hysteroscopy typically does not require the use of an electrical energy source, many options are available for the patient and surgeon.
Gas distension media
The only gas used in diagnostic hysteroscopy is carbon dioxide (CO2). Carbon dioxide is a safe distension media when used at pressures below the mean arterial pressure (MAP) of approximately 100 mmHg and flow rates less than 100 mL per min. The advantages of CO2 include high solubility in blood, high image clarity, and ubiquity in all operating rooms equipped for laparoscopy. Its use for hysteroscopy is limited to diagnostic cases and has significant disadvantages: the combination of intrauterine blood and CO2 creates bubbles that can dramatically reduce the viewable area and limit the diagnostic survey. The use of CO2 in hysteroscopy also requires the use of a hysteroinsufflator.
See the image below.
Laparoscopic insufflation devices are absolutely contraindicated because they cannot provide low rates of flow (30-100 mL per min) and safely limit the intrauterine pressure to less than 100 mmHg. Unintentional use of laparoscopic insufflators can cause CO2 embolisms that have led to cardiac arrhythmias and death. [30, 31, 32, 33]
Fluid distension media
The main advantage of fluid over gas is its ability to flush blood and tissue out of the viewing field. Fluid media can be functionally categorized into 2 groups: electrolyte and nonelectrolyte. The choice of media should take into consideration the patient comorbidities and the possibility of adding an operative component to the diagnostic procedure.
Electrolyte-containing solutions include normal saline (NS) and lactated Ringer’s solution (LR). Both are low viscosity fluids that are capable of conducting electric current. Neither can be used in procedures requiring monopolar energy. A recent systematic review found no difference in pain between the use of CO2 and NS as distension media. However, patients receiving NS experienced fewer vasovagal episodes, less shoulder pain, and had shorter operative times. [34, 35]
Authors of this and other studies recommend the use of normal saline for diagnostic hysteroscopy to prevent the need to change media should an operative procedure be required. [36, 37] Advantages of electrolyte solutions include compatibility with mechanical, laser, or bipolar energy sources and decreased concern for electrolyte imbalance from fluid extravasation. Disadvantages of electrolyte solutions include their miscibility with blood and their electrical conductivity.
Nonelectrolyte solutions include 1.5% glycine, 3% sorbitol, 5% mannitol, and 5% dextrose. With the exception of isotonic 5% mannitol, these low-viscosity solutions are hypotonic and nonconductive. Unlike NS and LR, they are all compatible with monopolar energy systems. The hypotonicity of these solutions imparts a risk of hyponatremia when absorbed in large volumes. Fluid deficits above 1500 mL or a serum sodium of less than 125 should prompt the immediate termination of the procedure and serial monitoring of postoperative blood chemistries. 
Glycine is metabolized to ammonia and should not be used in patients with significant hepatic dysfunction. Sorbitol is quickly metabolized into fructose and glucose and should be avoided in patients with impaired glucose tolerance. Evidence has suggested that 5% mannitol may be the safest of all nonelectrolyte solutions given its ability to remain in the extracellular compartment and maintain serum osmolality in the setting of concurrent hyponatremia. 
High-viscosity fluids are used in hysteroscopy to limit image distortion due to intrauterine bleeding. Currently, the only high-viscosity fluid option is Dextran 70. Dextran 70 is void of electrolytes and is therefore nonconductive.
Dextran 70 has been implicated in cases of pulmonary edema, disseminated intravascular coagulation (DIC), and anaphylaxis and should be used with caution. For every 100 mL absorbed, the plasma volume can expand an additional 860 mL.  Because intrauterine Dextran 70 absorption can occur after administration of volumes as low as 50-100 mL, actively monitoring fluid deficit and immediately terminating the procedure if the deficit reaches 500 mL is important. [41, 42] Care must be taken to wash the devices in hot water immediately after the surgery to prevent damage to the instrument by hardened Dextran. [38, 39]
Intraoperative fluid management is an important aspect of hysteroscopy. Both the American Congress of Obstetricians and Gynecologists (ACOG) and the American Association of Gynecologic Laparoscopists (AAGL) recommend the use of an automated fluid management system. [43, 44] These systems provide real-time information about the fluid deficit and can also actively manage intrauterine pressures. A member of the operating room personnel should be designated to call out the deficit when it reaches 500 mL and every additional 100 mL. With the exception of Dextran 70, the AAGL recommends that the surgeon should plan for completion of the case if the fluid deficit reaches 750 mL. The procedure should be immediately terminated if deficits reach 1500 mL for nonelectrolyte solutions or 2500 mL for isotonic solutions like NS or LR. If high volume deficits are suspected, electrolytes should be measured and diuretics given as indicated. 
Of note, hysteroscopic systems with ports for both inflow and outflow are superior in recording accurate fluid levels and decrease the amount of fluid left unaccounted on the table, drapes, and floor. If automated systems are unavailable, the distention fluid media can be administered and measured manually. Infusion pressure can be generated by elevating the fluid bag or by using an automated pressure infuser or blood pressure cuff. Pressure devices may increase fluid absorption and should be kept less than the patients MAP, especially during prolonged cases. In cases in which visualization is not optimal, the intrauterine distending pressure may be increased 5-10 mmHg above the MAP for short periods of time. Prior to use, the fluid should be warmed to a minimum of room temperature, as refrigerated fluids have been known to precipitate arrhythmias. 
Diagnostic hysteroscopy does not require operative instrumentation except for the potential use of graspers for foreign body removal and electrocautery for hemostasis.
Proper positioning is vital to the success of hysteroscopy. The patient should be prepared and draped in the dorsal lithotomy position with her legs in adjustable stirrups. Aseptic technique should be observed throughout the entirety of the case. A preoperative bimanual examination should be performed to ascertain the position and mobility of the uterus. Doing so increases the surgeon’s understanding of the anatomy and potentially decreases the likelihood of perforation. Trendelenburg positioning should be avoided during hysteroscopy because it may increase the risk of gas or air embolism, as discussed in the Complications section.
Once a patient has been selected for hysteroscopy, a history and physical examination should be performed to evaluate comorbidities. The patient should be up to date on cervical cytology screening. Cervical testing for Neisseria gonorrhea and Chlamydia trachomatas should be considered based on patient risk factors. The American Congress of Obstetricians and Gynecologists does not recommend the use of prophylactic antibiotics for hysteroscopy. 
Some authors have recommended monitoring of circulating oxygen levels with pulse oximetry during diagnostic office hysteroscopy to identify a gas or air embolism.  The cost effectiveness of such a recommendation may be questioned given the very rare occurrence of such an incident during a brief diagnostic procedure.
After informed consent, the patient is taken to the procedure room and placed in the dorsal lithotomy position. Of note, the consent form should include the possibility of a diagnostic laparoscopy in the event of a complicated uterine perforation. Once in the procedure room, a “time out” procedure is performed with the medical and nursing staff confirming surgical aspects including patient identity, procedure, signed consent, and drug allergies. The patient should have voided prior to the procedure, or a catheterization of the bladder can be performed following the surgical preparation.
After a bimanual examination, a bivalved or weighted speculum is used to bring the cervix into view. The cervix should be cleaned with an antiseptic solution. We recommend the use of 10% povidone-iodine or 4% chlorhexidine gluconate solution. A single-toothed tenaculum is then applied to the anterior lip of the cervix. A small amount of local anesthetic can be used prior to applying the tenaculum. If necessary, administer the paracervical block with the cervix under gentle traction.
In order to achieve an adequate paracervical block for the nonpregnant female, 10-20 mL of anesthetic should be injected 10 mm deep at the 4 o'clock and 8 o’clock position in the cervicovaginal junction with a 25-gauge spinal needle. Less commonly, injections are made at the 3, 5, 7, and 9 o'clock positions, but the increased number of sites does not improve analgesia and may disrupt the cervical branches of the uterine artery.  Another option is the intracervical block, which involves injections at the 2, 4, 8, and 10 o’clock or 3, 6, 9, and 12 o'clock positions of the cervical stroma and has been shown to control pain similarly to paracervical anesthesia.  As with all local anesthetic techniques, the site should first be aspirated and the injection made while slowly withdrawing the needle to avoid intravenous injection.
Because diagnostic hysteroscopy is generally a short procedure with minimal blood loss, CO2, NS, and LR are all appropriate distention media choices. Once equipment for distention media is activated and functional, the flow of the medium can be started. The tubing must be flushed with either CO2 or liquid prior to insertion into the cervix. As the hysteroscope is introduced to the external cervical os and advanced into the endocervical canal, attention should be turned to the video monitor or eyepiece. The distal tip of the hysteroscope is then gently advanced through the length of the cervix, taking care to keep the endocervical canal central on the viewing field if using a 0º scope.
Maintaining visualization of the endocervical canal when applying forward pressure to prevent cervical lacerations is important. Continuous gentle counter-traction via the tenaculum while advancing the hysteroscope allows the instrument to slowly slide into the uterine cavity. The flow of media should function to "wedge" a path to uterine entry.
Some advocate vaginoscopic “no-touch” hysteroscopy as a more painless technique. Such an approach forgoes the speculum and tenaculum and begins with vaginal and cervical disinfection followed by the intravaginal introduction of a rigid hysteroscope. The external cervical os is identified using the instrument light, and the examination is conducted as usual. Many authors have reported satisfactory results and good pain tolerance with this approach. [49, 50, 51, 52]
If a liquid distention media is used, intrauterine placement and orientation can be confirmed by visualizing bubbles at 12 o’clock. Once the distal tip of the hysteroscope is just inside the uterine cavity, allow the media to expand the intrauterine space. Adjustments to the rates of media inflow and outflow are done to expand the cavity, optimize image quality, and create smooth laminar flow.
The intrauterine space is then ready for systematic inspection. The first evaluation should be a panoramic view of the intrauterine cavity. Next, careful inspection of the following areas should be done: lateral uterine walls, superior uterine cavity, and anterior and posterior uterine walls. Gentle movement of the hysteroscope is imperative. Excessive trauma to the endometrial surface causes bleeding that obscures the view, increases systemic fluid absorption, and risks perforation. Any pathology should be inspected and documented.
If the hysteroscope lens is greater than 0º, twisting the bridge and barrel of the hysteroscope while keeping the camera perpendicular to the floor enables wide-angle viewing of the tubal ostia and other lateral structures. Flexible hysteroscopes have controls to maneuver the tip of the telescope. Pictures of the endocervical canal, pathology, and bilateral tubal ostia may be taken if desired.
See the images below.
Removal of a retained intrauterine contraceptive (IUC) is a common indication for hysteroscopy. For a description of this procedure, see the Medscape Reference article Operative Hysteroscopy.
If hysteroscopy is being performed to evaluate endometrial polyps or other intrauterine pathology, tissue samples can be obtained with hysteroscopic forceps followed by a dilation and curettage (D&C) after hysteroscopy. The hysteroscopy can be repeated after the D&C to confirm complete tissue removal. When the diagnostic survey is complete, the hysteroscope is slowly withdrawn. Careful inspection during removal provides one final chance to inspect the endocervical canal. A complete diagnostic hysteroscopy should take approximately 10-15 minutes.
Recovery after diagnostic hysteroscopy is prompt. If no anesthesia is used, patients can return to their usual diet and activities later that operative day. Some mild postoperative bleeding is normal and typically stops within 2-3 days.
The video below depicts diagnostic hysteroscopy used to evaluate abnormal uterine bleeding. The diagnosis was endometrial cancer.
Historically, hysteroscopy has been performed in the operating room setting, especially when high levels of patient anxiety and extensive operative procedures are anticipated. However, office hysteroscopy is becoming increasingly common because of smaller-diameter scopes and improved anesthetic techniques. If practitioners choose to implement office-based hysteroscopy, the procedures should be kept brief and should consist of either diagnostic or minor operative procedures. Prior to introducing an invasive technology into the office setting, it is important to implement and maintain rigorous safety measures as outlined by the American College of Obstetricians and Gynecologists’ Patient Safety Task Force report. 
Occasionally, diagnostic hysteroscopy may be used to locate and remove a foreign body. When doing so, obtaining preoperative imaging is prudent to rule out the possibility of migration out of the uterus and into the peritoneum, thereby necessitating laparoscopy.
Timing of the procedure
Diagnostic hysteroscopy should be performed in the proliferative phase of the menstrual cycle after cessation of menstrual flow. The procedure is typically performed prior to cycle day 12 in ovulatory women not using contraception for the theoretical concern of affecting oocyte or embryo transport or implantation. Examinations during the secretory phase should be avoided to allow for optimal views of the cavity since secretory endometrium can mimic intrauterine pathology.
Pharmacologic endometrial preparation
Endometrial preparation is more typically performed prior to operative hysteroscopy but may be necessary prior to diagnostic hysteroscopy, especially in anovulatory patients. Pharmacologic thinning of the endometrium can be done with a course of progestins, oral contraceptives, danazol, or GnRH agonists. A recent study evaluated the use of oral desogestrel plus vaginal raloxifene for rapid endometrial preparation. [54, 55]
Use of a tenaculum
Placement of a tenaculum on the anterior cervix is often the first cause of pain during a diagnostic hysteroscopy. With the availability of small diameter flexible scopes, most diagnostic procedures can be performed without a tenaculum. In the event that a tenaculum is needed, pain can be decreased with a topical anesthetic or by infiltrating the area of placement with local anesthetic using a 22-gauge to 25-gauge spinal needle. How the tenaculum is placed may also affect pain sensation. The cervix has few pain receptors but abundant pressure receptors. Thus, applying the tenaculum slowly with the smallest “bite” of tissue necessary diminishes discomfort. 
Hysteroscopy is a safe and well-tolerated procedure. The following complications are rare, but if left unrecognized can become life threatening.
Serum chemistry disturbances
CO 2 and air embolism
A recent study of over 13,600 diagnostic and operative hysteroscopies illustrates the overwhelming safety of diagnostic hysteroscopy. For diagnostic procedures, the overall complication rate was 0.28%. Uterine perforation was the most common complication (0.13% for diagnostic; 0.76% for operative).  This and other studies have shown that most uterine perforations occurred at the start of the case during early manipulation of the uterus and cervix and commonly occur during uterine sounding, cervical dilation, or hysteroscopic entry. Fluid overload did not occur in diagnostic procedures and was only reported in operative cases. [58, 59, 60]
Practitioners must understand the management of uterine perforation, which has repeatedly been identified as the most common complication. Rates of perforation can reach as high as 3% for operative hysteroscopic procedures and can lead to severe complications like bowel injury. [57, 58, 61] If a uterine perforation is recognized, all surgical instruments should be removed and the patient’s hemodynamic status assessed.
The most common site of perforation is the uterine fundus, which rarely leads to significant morbidity unless electrosurgical instrumentation is used. Injuries to this area can be managed conservatively because, generally, minimal bleeding occurs. Lateral perforations can lacerate the uterine vessels and can cause either immediate hemorrhage or the development of a retroperitoneal hematoma that can initially go unrecognized. A perforation in the cervix can injure the descending branches of the uterine artery and can also cause delayed hemorrhage. Note that delayed bleeding may be a sign of uterine or cervical perforation due to vessel spasm. An American College of Obstetrics and Gynecology Technology Assessment published in 2011 suggests the use of concurrent transabdominal ultrasonography during difficult cases in which perforation may be more likely. 
Management of uterine perforation or visceral injury is not within the scope of this article, but surgeons need to be able to recognize signs of organ injury, which include the following:
Acute onset of intraoperative or postoperative hematuria
Large and sudden fluid deficit
Excessive, uncontrollable bleeding
Bowel or omentum in the uterine cavity
Adipose tissue seen on pathologic specimen
Severe, persistent or unexpected pelvic/abdominal pain 
Heavy vaginal bleeding
Technical difficulties are rare but may include the inability to pass the hysteroscope due to severe cervical stenosis or unresectable anatomy that obstructs the intrauterine view. In addition to the above list, drug reactions to local anesthetics are no different than in other surgical procedures using similar methods and medications.
Air or gas embolism
Venous air or gas embolisms are rare complications of hysteroscopy. Air embolisms originate from room air and contain mostly oxygen and nitrogen. Because of nitrogen’s low solubility in blood, a lethal dose of room air is 5 times less than that of CO2. Room air can enter the circulation when the cervix is left open to room air or when an excessive amount of bubbles are left in the fluid system. Air embolisms can also occur as a result of repetitive instrument insertions through the cervix causing a “pistonlike” transmission of air into the uterus.  Steep Trendelenburg positioning may also increase the risk of embolism as a negative pressure gradient is created from the genital tract to the heart. 
Carbon dioxide embolisms causing cardiopulmonary decompensation resulting from hysteroscopy are a rare and serious complication of CO2 use. These gas embolisms occur when large amounts of CO2 interface with damaged blood vessels inside the uterus. Inadequate purging of air from the delivery tubing may also play a role. 
Entry of air into the circulation may actually not be uncommon during hysteroscopy. In a study of 23 women undergoing operative hysteroscopy under general anesthesia, bubbles within the right atrium were detected by transthoracic echocardiography in all subjects. Eighty-five percent of the subjects showed a continuous flow of bubbles during the procedure.  Interestingly, a study confirming this also showed that the liver can trap venous gas as it passes through the inferior vena cava, suggesting that symptoms of embolus only occur when large volumes of gas overwhelm the hepatic bubble trap. 
Symptoms of air or CO2 embolism include chest pain, dyspnea, and decreases in oxygen saturation or end tidal CO2. A loud, machinelike heart murmur, or “mill wheel murmur,” can occasionally be heard due to the mixing of blood and air in the right ventricle. Wheezing and rales can occur as a result of acute bronchospasm and pulmonary edema. If an embolus is suspected, the patient should be administered 100% oxygen and quickly repositioned.
Placing the patient in a combined Trendelenburg and left lateral position (Durant’s maneuver) may permit the entrained air bubble to move to a nonobstructing position inside the right ventricle. [33, 69, 70] The withdrawal of venous air can be performed with central venous catheters, and in the case of cardiac arrest, direct needle puncture of the right heart has been advocated to sustain circulation. [33, 65, 66, 71, 72] One can theoretically decrease the possibility of both air and gas embolus by minimizing the exposure of the open cervix to room air and avoiding steep Trendelenburg positioning during the procedure.
Note the following:
Hysteroscopy is a safe and well-tolerated procedure that is performed for diagnostic or operative indications.
Diagnostic hysteroscopy is indicated in cases of abnormal uterine bleeding, suspected Müllerian anomalies, removal of foreign bodies, and when abnormal imaging findings need to be confirmed.
Diagnostic hysteroscopy cannot be performed in the setting of a viable intrauterine pregnancy, cervical or uterine cancer, active pelvic infection, or when the primary surgeon is inexperienced.
Pregnancy testing should be performed in the premenopausal woman prior to hysteroscopy.
Anesthetic options for women undergoing hysteroscopy are highly varied and range from nothing to general endotracheal anesthesia. The pain during the procedure is usually proportional to the outer diameter of the hysteroscope and amount of cervical dilation.
Flexible diagnostic hysteroscopes less than 5.0 mm in diameter do not usually require anesthesia or any mechanical dilation of the cervix. For this reason, diagnostic hysteroscopy is generally well tolerated in the outpatient setting.
Common distension media for diagnostic hysteroscopy are normal saline and CO 2 gas. Isotonic solutions like normal saline and lactated Ringer’s solution are frequently used because of their availability, safety, and ability to flush out blood and particulate matter. Isotonic solutions are not compatible with monopolar energy.
Hypotonic nonelectrolyte solutions (1.5% glycine, 3% sorbitol, 5% mannitol, 5% dextrose in water) are compatible with all types of energy sources and are less frequently used in diagnostic hysteroscopy.
For the premenopausal woman, diagnostic hysteroscopy is best performed during the early proliferative phase of the menstrual cycle. For patients with significant endometrial tissue, pharmacologic thinning of the uterine lining may be indicated.
Complications arising from diagnostic hysteroscopy are rare but can become serious if left unrecognized. The most common complications are uterine perforation and bleeding. Other more rare and morbid complications are air and CO 2 embolisms, postoperative infections, and severe electrolyte disturbances.