Jon I. Einarsson, MD, PhD, MPH

home

authors

jon-i-einarsson-md-phd-mph

Artificial intelligence (AI) is already all around us in our daily lives. It enables smart phones to sort our pictures based on face recognition algorithms, it helps Google guess what you‘re going to search for the second you start typing and it helps Uber figure out what you are willing to pay for various services. 


	AI is also entering medicine. That is, in part, enabled by the large amount of high-resolution digital data that are collected in electronic medical records and from images, genetic screening, and laboratory tests. The other vital component is the ever-increasing capacity of computers to analyze large amounts of data and create a solution on their own, a process called machine learning. Once implemented, machine learning algorithms can continuously improve their performance by using outcomes to improve the fit of the model. These methods have the potential to help with diagnostic accuracy, outcome prediction, cost-effectiveness assessment, and real-time decision-making in surgery.

Can AI improve accuracy and efficiency of mammography?

Han et al discuss use of AI in ob/gyn ultrasound

 As they point out, AI is especially helpful in image detection and automation. It has the potential to reduce clinician errors and help detect rare anomalies, especially among less skilled providers, thereby reducing clinician variability in terms of performance. AI also has a lot of other potential applications in our field and in all of medicine. A recent systematic review identified nine articles that spanned multiple medical subjects, including retinal diseases, skin cancers, pulmonary nodules, and brain tumors.

All of these articles compared the performance of clinicians and AI and found a consistent theme. Current AI algorithm development has a diagnostic performance comparable to that of medical experts, especially in image recognition-related fields. This is especially impressive, given how fairly recently AI has been applied to medical decision-making. It seems inevitable that this technology will improve rapidly over the next few years, which should improve accuracy significantly. 


	This does not mean that clinicians are not needed anymore. AI is not meant to replace clinicians (at least not yet), but rather, to augment their diagnostic and therapeutic capacity. One of the major concerns for using AI in medicine is the lack of transparency about how decisions are generated and the inherent risk of bias if the training database is skewed. Some recent advances in the field aim to mitigate those risks. For example, some of the newest diagnostic machine learning algorithms are able to create saliency maps, which identify the parts of  the image that contributed the most to the diagnosis. 

These AI-powered modalities will increasingly help physicians in their daily tasks in the clinic and lab and also potentially in the operating room. Because most surgeries are now image-based, i.e., either laparoscopic or robotic, there is ample opportunity to use neural networks and machine learning to identify patterns in how surgeons perform procedures and potentially alert them if they are diverting off course. It seems likely that this technology will be incorporated into surgical suites in the not-so-distant future. The increasing incorporation of AI into medicine has the promise to improve patient care and patient outcomes and it merits careful exploration and study. 

Datkhaeva I, Han CS. Artificial intelligence in ob/gyn ultrasound. Contemp Ob/Gyn. 2019;64(10):XX.

Shen J, Zhang CJP, Jiang B, et al. Artificial intelligence versus clinicians in disease diagnosis: systematic review. JMIR Med Inform. 2019 Aug 16;7(3):e10010.

Articles

As artificial intelligence increases its presence in medicine, ob/gyns are welcoming the technology with both excitement and trepidation.

AI: Coming to a clinical scenario near you

 Jon I. Einarsson, MD, PhD, MPH, and Nisse V. Clark, MD, discuss their novel technique for Essure reversal in this video. To read their full article, click 

Videos

Jon I. Einarsson, MD, PhD, MPH, and Nisse V. Clark, MD, discuss their novel technique for Essure reversal in this video.

Video: Essure Removal Tips

For over a decade, transcervical sterilization revolutionized gynecology by offering an office-based procedure for sterilization under local anesthesia. Essure® was the most successful hysteroscopic sterilization device on the market, approved for almost 16 years in the United States and with an estimated 750,000 devices sold worldwide.

While Essure proved to be very effective in preventing pregnancy, recent years have seen a dramatic rise in the number of adverse events (AEs) associated with it reported to the US Food and Drug Administration (FDA).

Several countries have withdrawn the product from the market and Bayer AG recently announced it will discontinue sales in the United States by the end of 2018.

A growing number of women are now seeking surgical removal of Essure due to suspected AEs such as pelvic pain, heavy bleeding, allergic reaction, and a multitude of other symptoms that may be related to the device.

Essure removal is typically performed laparoscopically by removing both fallopian tubes as well as the interstitial portion of the microinsert. Many women also choose to undergo a cornual wedge resection or hysterectomy to ensure that the device and any fragments are completely excised. 

Less commonly discussed is the issue of sterilization regret, a condition that affects up to 20% of sterilized women and likely some women with Essure.

Traditionally, Essure sterilization is considered irreversible due to extensive scarring along a large portion of the fallopian tube. Most patients desiring pregnancy after Essure are advised to undergo in vitro fertilization (IVF), a readily-available alternative that has shown some success with Essure in situ.

But what about women who desire removal of the device and restoration of fertility? For them, a procedure that removes the device and offers even a small chance of spontaneous pregnancy might be worthwhile. 

Essure reversal is uniquely challenging. The device’s polyethylene terephthalate fibers induce permanent fibrosis along almost half of the fallopian tube, including the interstitial segment. This leaves only a short segment of the distal fallopian tube available for direct attachment to the uterus. This type of procedure is termed a tubouterine anastomosis; a technique that dates back to the late 19

Tubouterine anastomosis was first described for treating women with proximal tubal occlusion unrelated to sterilization.

Proximal tubal occlusion can be caused by conditions such as chronic salpingitis, salpingitis isthmica nodosa, cornual fibroids, adhesive disease, and endometritis.

Long before IVF, tubal cannulation and tubouterine anastomosis were attempted to restore tubal patency in women with proximal tubal occlusion. Tubouterine anastomosis was classically an open procedure in which fundal hysterotomy or cornual wedge resection was performed for excision of the affected fallopian tube and reimplantation of the distal fallopian tube to the uterine cornua. Conception rates following the procedure were variable, ranging from 14% to 42% in the literature.

Montieth et al. are the only authors to publish their experience using tubouterine anastomosis to reverse Essure sterilization. Their technique uses a 5-to 10-cm laparotomy and transfundal posterior uterine incision or cornual wedge resection to reimplant the distal fallopian tubes into the uterine cornua. In a published series of 70 cases, the 12-month post-procedural conception rate was 36%.

A more up-to-date analysis on their website reports a 38% pregnancy rate  for 282 procedures performed over 9 years.

Also reported is a 5% ectopic rate and 4% uterine rupture rate. 

We have tested a new method for Essure reversal using a laparoscopic approach that does not require a large abdominal or uterine incision. With this method, laparoscopy and hysteroscopy are performed simultaneously to remove the device and reimplant the distal fallopian tube into a new site on the uterine fundus. We present a description and video of our technique as well as our experience with three cases.

The microinserts are identified and laparoscopic scissors are used to partially transect the fallopian tube at the distal end of the device (Figure 2). Electrosurgery is avoided to preserve as much healthy fallopian tube as possible. Once the distal end of the Essure device emerges, it is grasped laparoscopically and pulled out of the fallopian tube (Figure 3). The device uncoils and elongates as it is removed and can easily fracture. Pulling the device parallel to the tube and intermittently regrasping it near its exit from the tube can help avoid device fragmentation.

Simultaneous hysteroscopic guidance is used to ensure that the intracavitary portion of the microinsert is completely removed. We have found that a fractured device may appear to be completely removed laparoscopically, when in fact, a small portion of the proximal microinsert remains visible hysteroscopically. 

Once both Essure microinserts are removed, semi-rigid stents are inserted into the tubal ostia through the operative channel of a 5-mm hysteroscope (Figure 4). Despite insertion of the stents into the tubal ostia, the stents do not emerge from the proximal fallopian tube because of the fibrosis caused by Essure. In our experience, the device instead perforates the uterine fundus slightly posterior and medial to the tubouterine junction. Once the stent emerges, it is cannulated into the distal fallopian tube segment laparoscopically.

The distal fallopian tube is brought to the uterus along the stent and the anastomosis performed. Interrupted stitches are placed circumferentially around the new tubouterine junction with knots tied intracorporeally (Figure 5). Stitches are placed until the fallopian tube is well-approximated to the uterus and the stent is no longer visible. The process is repeated on the other side and the stents are removed. At the end of the procedure, chromopertubation is used to assess tubal patency (Figure 6).

Three laparoscopic Essure reversal procedures were performed at Brigham and Women’s Hospital between 2017 and 2018 using this technique (Table 1). All three patients were in their 20s when they underwent Essure sterilization and later regretted the procedure. All were experiencing side effects of the Essure device and had a history of spontaneous pregnancies prior to sterilization. The three women were 27 to 35 years of age at the time of Essure reversal. 

Operative time ranged from 87 to 142 minutes and estimated blood loss was minimal at 20 to 75 mL. All procedures were uncomplicated and patients were discharged home on the same day as surgery. A follow-up hysterosalpingogram was done 1 to 3 months following the procedure and showed bilateral patency in one case and unilateral patency in the other two cases. While restoration of tubal patency is promising, no pregnancies have occurred over a follow-up period of 5 to 11 months. If the patients are able to conceive, risk of uterine rupture may be reduced by avoiding a hysterotomy or cornual wedge resection. 

IVF is still a good option for most women who desire pregnancy after Essure, especially if the device is not causing symptoms. Literature on IVF for Essure regret is limited to a report of two cases, both of which resulted in a live birth.

Several small studies have also evaluated intentional placement of Essure prior to IVF for infertile women with hydrosalpinx.

In this setting, the microinserts occlude the proximal fallopian tube and prevent hydrosalpingeal fluid from spilling into the uterine cavity. A systematic review of Essure and IVF for infertile women with hydrosalpinx reported a 29% ongoing pregnancy and live birth rate.

Presumably, IVF success rates for Essure regret are greater than those in women with underlying infertility.

When weighing the risks and benefits of IVF compared to Essure reversal, it is worth considering what we know about IVF and tubal reanastomosis after tubal ligation. As with any fertility procedure, tubal reanastomosis after tubal ligation is more likely to benefit younger women.

A retrospective study by Boechxstaens et al. found that cumulative delivery rates were significantly greater after tubal reanastomosis compared to IVF in women younger than age 37 (72% vs. 52% over 72 months).

A decision analysis by Messinger et al found that tubal reanastomosis was more cost-effective for women younger than age 41, whereas IVF was more cost-effective for women aged 41 and older.

Essure reversal is also assumed to be less advantageous for older women. 

Essure reversal with tubouterine anastomosis is technically and functionally a different procedure from tubal reversal with end-to-end anastomosis of the fallopian tube. The greater extent of tubal damage following Essure sterilization is very likely to result in lower success rates than other tubal reversal procedures. While laparoscopic Essure reversal can restore tubal patency, it is currently unknown if it can result in a normal pregnancy. Nevertheless, the opportunity to remove a device associated with symptoms may prompt younger women with Essure regret to attempt reversal despite presumably low success rates.

The authors report no potential conflicts of interest with regard to this article.

Bayer. Essure permanent birth control. http://www.essure.com/what-is-essure. Accessed August 15, 2018.

Walter JR, Ghobadi CW, Hayman E, Xu S. Hysteroscopic Sterilization With Essure: Summary of the U.S. Food and Drug Administration Actions and Policy Implications for Postmarketing Surveillance. Obstet Gynecol. 2017;129(1):10-19.

Mohan N. Bayer to stop selling Essure in the U.S. Wall Street Journal. July 20th, 2018.

Kamencic H, Thiel L, Karreman E, Thiel J. Does Essure cause significant de novo pain? a retrospective review of indications for second surgeries after Essure placement. J Minim Invasive Gynecol. 2016;23(7):1158-1162.

Clark NV, Rademaker D, Mushinski AA, Ajao MO, Cohen SL, Einarsson JI. Essure removal for the treatment of device-attributed symptoms: an expanded case series and follow-up survey. J Minim Invasive Gynecol. 2017;24(6):971-976.

Hillis SD, Marchbanks PA, Tylor LR, Peterson HB. Poststerilization regret: findings from the United States Collaborative Review of Sterilization. Obstet Gynecol. 1999;93(6):889-895.

Arora P, Arora RS, Cahill D. Essure® for management of hydrosalpinx prior to in vitro fertilization-a systematic review and pooled analysis. BJOG. 2014;121:527–536.

Kerin JF, Cattanach S. Successful pregnancy outcome with the use of the in vitro fertilization after Essure hysteroscopic sterilization. Fertil Steril. 2007;87:1212.e1-4.

Reis E. Plastic operation on the fallopian tubes. Am J Surg Gynecol. 1899;11:180.

De Silva PM, Chu JJ, Gallos ID, Vidyasagar AT, Robinson L, Coomarasamy A. Fallopian tube catheterization in the treatment of proximal tubal occlusion: a systematic review and meta-analysis. Hum Reprod. 2017;32(4):836-852.

Turck RC. Hysterosalpingostomy. NY Med J. 1909;89:1193-1196.

Williams GF. Tubo-uterine implantation with special reference to reversal of sterilization. Lancet. 1969;1(7599):825-827.

Green-Armytage VG. Tubo-uterine implantation. J Obstet Gynaecol Br Emp. 1957;64:47.

Diamond E. A comparison of gross and microsurgical techniques for repair of cornual occlusion in infertility: a retrospective study, 1968–1978. Fertil Steril. 1979;32:370-376.

Montieth CW, Berger GS, Zerden ML. pregnancy success after hysteroscopic sterilization reversal. Obstet Gynecol. 2014;124(6):1183-1189.

Tubal Reversal A Personal Choice. https://www.tubal-reversal.net/. Accessed August 15, 2018.

Trussel J, Guilbert E, Hedley A. Sterilization failure, sterilization reversal, and pregnancy after sterilization reversal in Quebec. Obstet Gynecol. 2003;101(4):677.

Boeckxstaens A, Devroey P, Collins J, Tournaye H. Getting pregnant after tubal sterilization: surgical reversal or IVF? Hum Reprod. 2007;22(10):2660.

Messinger LB, Alford CE, Csokmay JM, Henne MB, Mumford SL, Segars JH, Armstrong AY. Cost and efficacy comparison of in vitro fertilization and tubal anastomosis for women after tubal ligation. Fertil Steril. 2015;104(10):32-38. 

With the impending withdrawal of Essure from the US market, ob/gyns should know how to remove the device. The authors describe the traditional technique and their new approach.

Laparoscopic Essure tubal reversal: How we do it

As gynecologic surgery continues to evolve rapidly, it is easy to forget the origins of our specialty. Only 205 years ago, Conrad Langenbeck performed the first planned hysterectomy on record. It was a vaginal procedure done in the surgeon‘s living room with no assistants and he had to use his teeth at one point to tie suture. Miraculously the patient survived, but no one believed that Dr. Langenbeck actually performed this procedure until it was confirmed at autopsy several years later. 

In today‘s environment, the laparotomy approach is still overutilized, but it is now gradually being replaced by laparoscopic and robotic surgery. The original minimally invasive method, the vaginal hysterectomy, is slowly on the decline despite efforts by the American College of Obstetricians and Gynecologists to turn the tide. There is no question that vaginal hysterectomy is a great minimally invasive approach, but with fewer surgeons capable of teaching this skill and a paucity of technological innovation in this space, vaginal hysterectomy has been left wanting compared with the laparoscopic approach. In this regard, other developments on the horizon, which I will discuss later, have the ability to further cement the laparoscopic minimally invasive visual approach as the primary mode of access for gynecologic surgery.

The impact of MIGS training and subspecialists

As compared with vaginal hysterectomy, one important factor tilting the scales is development of minimally invasive gynecologic surgery (MIGS) fellowship programs. Over 300 MIGS-trained subspecialists have graduated from fellowship since 2001, and every year 30 to 40 new MIGS-trained subspecialists pollinate the US landscape and start to operate on patients. Introduction of high-volume MIGS subspecialists is changing the landscape of gynecologic surgery, moving a large portion of the surgical volume away from generalist specialists to MIGS specialists. A similar evolution happened in gynecologic oncology a couple of decades ago with the advent of gynecologic oncology fellowships. Whether this is the ideal solution for the discipline as a whole can be argued, but it is likely that within the next 10 to 20 years, most complex benign gynecologic surgeries will be performed by subspecialists and several recent studies have demonstrated in multiple arenas that high-volume surgeons have better outcomes and fewer complications.

Another exciting development is a new revolution in robotic surgery that we are about to witness. Until recently, Intuitive Surgical was the only game in town, but recently the Senhance surgical robotic system was launched in the United States. Other robotic systems are on the horizon, some from major players in the market such as Medtronic and J&J in collaboration with Google, but also a variety of systems from other companies. It seems evident that with increasing competition, innovation will be rapid and prices will come down. This will benefit patients and surgeons and reduce the overall cost of the healthcare system. While I have personally been critical of robotic surgery as a poor value proposition in its current iteration, I now see the possibility that in time, robotic surgery will be offered at a signficantly lower cost and offer true value such as automation and greater safety for patients. 

Augmented visualization is also going to be something we will see increasingly in the near future.

Fluorescence-guided surgery is already in clinical use, but several companies are developing tools that incorporate augmented reality in which three-dimensional images from computed tomography or magnetic resonance imaging can be overlaid onto the surgeon‘s console or monitored in real time. This can help the surgeon better determine exactly where to make an incision and see healthy tissue margins. There are also several dyes and imaging enhancements being developed for display of various tissue types on the monitor in distinct colors. In time, a surgeon will be able to see the anatomy almost as if it were in an anatomy atlas, with nerves, ureter, bowel, and muscles all in different colors on the monitor. Essentially, it will be a kind of a surgical GPS that guides the surgeon to address the correct pathology while avoiding injury to vulnerable nearby structures. 

Continued innovation is the driver of this progress. Large multinational medical device companies lead the way in development of novel medical devices and equipment, but the importance of small startups should not be underestimated. Small startups are generally run by passionate and knowledgable founders who have identified a significant unmet need and are able to move quickly through development of their devices because of their small organizational structure and the unified drive of the group. The main challenge for small startups is access to funding and it is well recognized that most of these companies will ultimately fail, often because they run out of money. However, there are inspiring success stories. One is Surbhi Sarna (now 32 years old) who was diagnosed at age 13 with ovarian cysts. Her doctors did not know if the cysts were cancerous, which inspired Ms. Sarna to found a company, nVision Medical. The company raised $17 million in venture capital and received approval from the US Food and Drug Adminstration (FDA) to develop a specialized device to detect ovarian cancer before being aquired by Boston Scientific for $275 million. It is obvious that Ms. Sarna is a very gifted individual to have accomplished this at such a young age, but it is nevertheless a good example of the power of the individual entrepreunial spirit.

The regulatory environment for medical devices has recently come under increased scrutiny, such as is shown in “The Bleeding Edge.” In this documentary, the medical device industry is painted in a rather negative light, with the focus on (among other things) issues that are directly pertinent to women’s health, such as Essure sterilization implants, vaginal mesh, and robotic surgery. The documentary highlights some potential problems with the current regulatory system for medical devices. The 510(k) pathway enables medical device companies to bring a device to market based on a similar device that is already on the market, called a predicate device. The problem is that sometimes the approval of that predicate device is based on a string/hierarchy of other predicate devices, meaning that the original predicate device is no longer on the market and in some cases may be outdated. 

On the other hand, the 510(k) pathway enables small startups to bring their innovations to market without going through the premarket approval (PMA) process, which includes a clinical trial, a usually cost prohibitive step. In addition, even if a clinical trial is done, it may not necessarily prevent injury or issues for patients later on. An example of this is the Essure sterilization system. The device did go through the PMA process with a rigorous clinical trial, but because of the relatively short follow-up period, the problems associated with Essure were not discovered until several years after the clinical trial was completed. 

The FDA’s current system for medical device approval is designed to balance the need to protect patients while continuing to encourage innovation. One obvious downfall is that the FDA’s MAUDE (Manufacturer and User Facility Device Experience) database relies on voluntary reporting from doctors, nurses, and patients while reporting is mandatory for manufacturers, importers, and device-user facilities. Consequently, there is a potential for underreporting and there is a risk for late detection of problems with medical devices, ie the issue is not detected by the FDA or authorities until there has been significant patient harm. There seems to be a need for a more comprehensive nationwide database to protect patients, wherein medical device complications are registered prospectively and registration is mandatory by all parties. 

In summary, innovation in surgery is necessary if we wish to see improvements in patient safety and overall care. However, ideas alone are not the sole component. Innovation requires skilled surgeons willing to teach, less-skilled surgeons willing to learn, and a regulatory environment that properly protects patients without barriers that are too high to allow technology to evolve. ï¼

Innovation is a major component of gynecologic surgery and surgeons at all levels play a role in progressing the specialty

Author | Jon I. Einarsson, MD, PhD, MPH

Articles

AI: Coming to a clinical scenario near you

Video: Essure Removal Tips

Laparoscopic Essure tubal reversal: How we do it

Gynecologic surgery’s future