While physicians have peered into the depths of the human body through its natural openings for more than 100 years, natural orifice translumenal endoscopic surgery, or NOTES™, dates back to 2005. This occurred when a group of gastroenterologists and surgeons convened in an attempt to propagate this disruptive concept of minimally invasive surgery in a thoughtful, scientific, and safe manner. The meeting was catalyzed by a report of “flexible transgastric peritoneoscopy” published by Kalloo et al. in 2004. The procedure was performed in a swine model, and subsequent animal work by the same group at Johns Hopkins University demonstrated the feasibility of procedures such as transgastric ligation of fallopian tubes, cholecystectomy, gastrojejunostomy, and splenectomy [1]. The novel innovation was the use of the flexible endoscope as the operating platform.

The slow progress of utilizing a natural orifice has gone from simply looking to performing procedures adjacent to the opening, and finally to performing procedures far from the natural orifice. While all procedures were both enhanced and limited by one technological device or another, the technologic restrictions did not limit the imagination and foresight of the surgical and gastroenterological pioneers that laid the foundation for NOTES™ as we know it today.

In 2005, fourteen thought leaders, representing the Society of American Gastrointestinal Endoscopic Surgeons (SAGES) and the American Society of Gastrointestinal Endoscopy (ASGE), assembled in Phoenix, Arizona, to form a working group on this nascent field. The result of this meeting was an important white paper written by the working group published in 2006 [2]. There were three critical accomplishments from this meeting.

The first accomplishment was an agreement on nomenclature. Although the focus at the time of the meeting was on transgastric surgery, the leaders recognized that other routes of access to the abdomen, namely transvaginal or transcolonic, could also develop. The term “natural orifice translumenal endoscopic surgery” was adopted to describe this, and the acronym NOTES™ was born. It was also uniformly agreed upon that these were to be considered surgical procedures because “tissue resection and repair is the ultimate goal of accessing intraperitoneal organs.” The working group named itself the Natural Orifice Surgery Consortium for Assessment and Research (NOSCAR), a clever acronym for the development of incisionless surgery. While it may seem trivial to have spent so much effort on nomenclature and taxonomy, one only needs to consider the bewildering sea of names and acronyms created to describe techniques and devices used in single-port laparoscopic surgery to realize that agreement on nomenclature is important [3].

The second accomplishment was to define criteria by which one could participate in NOSCAR, with an eye on avoiding the large increase in complications caused by the last revolution in gastrointestinal surgery: the introduction of laparoscopy. In the name of patient safety, NOSCAR outlined the following criteria for participation:

The third accomplishment was to define the current limitations in the ability to perform NOTES™ procedures. They outlined the following eleven potential barriers to the safe introduction of NOTES™ in human patients:

This paper immediately set into motion those interested in NOTES™ research and development, as well as the clinical introduction of these techniques.

Transvaginal surgery dates back to ancient times. Some claim that the first surgical procedure described in history was a vaginal hysterectomy by Themison of Athens in 50 BC. Others claim the first vaginal hysterectomy was performed by Soranus of Ephesus in 120 AD, whose treatise, gynecology, has survived into modern times (translated into English in 1956) [4, 5]. This was considered a seminal work at the time and provided a look at ancient obstetric and gynecological techniques. Soranus described a transvaginal hysterectomy for severe uterine prolapse associated with ischemia and gangrene. Morbidity and mortality, however, were almost universal. In the eleventh century, an Arabian physician, Alsaharavius, wrote of vaginal hysterectomy, and there are some who believe that these patients survived. Clearer reports of survival date back about 500 years; Berengarius da Carpi of Bologna in 1507 performed a partial vaginal hysterectomy on a patient who survived. More incredible is the case of Faith Howard, a 46-year-old peasant, who performed a vaginal hysterectomy on herself in 1670. She was said to be carrying a heavy load when her uterus prolapsed completely. Frustrated by this frequent occurrence, she grabbed her uterus, pulled as hard as possible, and cut the whole lot of it with a short knife. The bleeding soon stopped and she lived on for many years, with a persistent vesico-vaginal fistula [5, 6]. The first elective cases began to appear in the literature from France and Germany shortly after 1800—and many years before Charles Clay reported the abdominal hysterectomy in 1843, which unfortunately was unsuccessful due to an incorrect diagnosis and early postoperative mortality [5, 7]. Vaginal hysterectomy for cancer was reported in 1892 by Terrier and Hartman [8].

In the late 1800s Durhssen, Mackerodt, and Martin of Berlin, Germany, utilized anterior colpotomy to perform transvaginal operations of the tubes, ovaries, and uterus for a variety of conditions, including ectopic pregnancy, and the use of “morcellement” for the removal of very large uterine myomas. At the annual meeting of the British Medical Association in London in 1895, Martin touted the decreased morbidity compared to laparotomy as justification for the approach. He closed the colpotomy initially with silk and silver wire, but abandoned these for juniper catgut. Postoperatively, the vaginal wound was said to take only 8–10 days to heal, “so that about the twelfth day the patient may be allowed to leave bed.” No local treatment was necessary, and all of his 152 cases recovered without “feverish reaction.” [9].

In 1901, Russian gynecologist Dmitry Von Ott presented his work in St. Petersburg using a posterior colpotomy for a variety of gynecological operations. Unique to his approach, which he dubbed “ventroscopy,” was the use of “a peanut-sized lamp and a spoon-shaped shield to protect the patient from burn,” and reflected light into the cavity using metallic mirrors and a headlamp. He also utilized the Trendelenburg position and placed a cotton swab in the vagina, allowing filtered air to be vacuumed into the peritoneal cavity, creating a “natural form of insufflation.” His technique was used by the Europeans into the 1920s to 1930s, and by the Americans from the 1940s to 1960s [10].

Transvaginal tubal ligation was further advanced by the Japanese in 1970 [11], and there were additional scattered case reports throughout the 1970s. Transvaginal oophorectomy (at the same time as a hysterectomy) was revisited by Nichols in 1978 [12]. He noted that access to the ovaries could be very challenging due to the constraints of the size of the colpotomy and the bony anatomy.

Transvaginal specimen extraction was first described in the early 1990s. Delvaux et al. from Brussels [13] described a case report of a laparoscopic cholecystectomy in a woman with large gallstones, where they opted for specimen removal via a posterior colpotomy rather than removal from a larger abdominal wall incision. Also in 1993, Breda et al. in Italy used a posterior colpotomy for extraction of a tuberculous left kidney after laparoscopic nephrectomy [14]. Although these reports were overlooked at the time, they were proof of concept that organs not immediately adjacent to the vagina can be safely removed via this natural orifice.

In the mid-1990s, there were two reports of transvaginal oophorectomy using an endoscope. In London Magos published a technique using a standard laparoscopic instrumentation without pneumoperitoneum, and Yuen in Hong Kong, in his published experience, commented on the difficulty of manipulating three instruments in such a small working space—a prescient statement given the subsequent development of single-site surgery and the challenges that the concept introduced [15, 16].

It was not until the turn of the twenty-first century, about 100 years since the first surgeon peered through the vaginal vault into the peritoneal cavity using a small lamp with metal mirrors that our current concept of transvaginal NOTES™ came into being.

In 2007, Scott et al. used a swine model to perform transvaginal cholecystectomy utilizing a proprietary magnetic instrument system [17]. The first report of a transvaginal cholecystectomy in the USA was presented as a video at the April 2007 annual meeting of the Society of American Gastrointestinal and Endoscopic Surgeons (SAGES). This seminal video was the first time a human NOTES™ cholecystectomy had been presented, and it utilized a hybrid approach with a transvaginal flexible endoscope and umbilical laparoscopic port and took approximately 3 h. This report was immediately criticized, with one surgeon interviewed for a New York Times article calling the procedure “repulsive,” stating that “the idea of puncturing internal organs and then sewing them up was cause for concern”. She further stated that “As a woman I find it very invasive, physically and emotionally. To me it’s quite distasteful.” [18]. The report of the video was published in December of the same year [19].

The first South American report of transvaginal cholecystectomy also appeared in the literature in 2007. Zorron et al. performed a case the same week as Bessler’s group in New York, and the case was successfully completed with endoscopic instruments placed alongside the vaginal endoscope. Amazingly, the procedure lasted only 66 min, and this technique quickly gained popularity in Brazil [20].

Also in 2007, Marescaux and colleagues at IRCAD (Institut de Recherche contre les Cancers de l’Appareil Digestif; French: Institute for Research into Cancer of the Digestive System) in Strasbourg reported a hybrid transvaginal cholecystectomy using a flexible endoscope aided by a single, 2-mm, laparoscopic port in a human. The authors were careful to report that “all of the principles of cholecystectomy were strictly adhered to,” and that the patient had an uneventful recovery [21]. A second human case, this one from Hamburg, was published a month later, where Zornig and colleagues performed a cholecystectomy with standard laparoscopic instrumentation (i.e., a rigid endoscope) via the vagina and one umbilical port. The authors emphasized that the vaginal access was better for closure and infection, and rigid instrumentation was easier to use than a flexible endoscope [22]. This approach became common in Europe for those who adopted the procedure.

The first case report of human transvaginal appendectomy came in 2008 from India. The report details six attempted cases: the first three failed, the next two were completed with a 3-mm laparoscope in the umbilicus, and the final case succeeded as a purely transvaginal case. In the final case, pneumoperitoneum was initiated via a transumbilical Veress needle. Hot biopsy forceps were used to divide the mesoappendix, and endoloops were used to secure the appendix. All patients in the series recovered uneventfully [23].

After these early reports there have been hundreds of articles published about transvaginal operations. While the vast majority of these focused on cholecystectomy, removal of all of the GU organs, liver, spleen, and stomach have also been reported. Exploration and specimen removal have also been reported, along with hernia repair.

Importantly, genuine concerns about the safety and appropriateness of a transvaginal approach gave rise to a host of published surveys given to patients, spouses, and healthcare workers. These surveys suggest there is no specific patient type that prefers a transvaginal approach over standard laparoscopy. Results are highly variable when examined by age, reason for operation, and BMI. Concerns regarding scarring and cosmesis were generally less important compared to issues related to safety and recovery time. In general, patients were more likely to accept a transvaginal approach compared to healthcare workers [24–33].

The first published cases of transgastric surgery appeared in 1950, when Scovel and Holliger reported a case of transgastric pancreatocystogastrostomy [34]. In 1959, Brewer and Shumway reported transgastric catheter drainage of a pancreatic pseudocyst [35]. Although both of these cases were performed via laparotomy, they were the first cases that described crossing the lumen of a hollow organ to gain access to the operative field, an important precursor to NOTES™. Another interesting concept using a transgastric approach was published by Petropoulos in 1979, where he described a transgastric route for highly selective vagotomy to control peptic ulcer symptoms [36].

In 1980, Ponsky et al. in Cleveland described the percutaneous placement of a gastrostomy tube, rather than placement via a laparotomy incision [37]. A monumental achievement at the time, this drastically reduced the invasiveness of gastrostomy tube placement. The procedure was accomplished via a natural orifice, traversed the lumen of a hollow viscus, and utilized a flexible endoscope. One could easily argue that Ponsky’s PEG tube is the very first NOTES™ procedure by today’s standards.

Another use of transgastric surgery was for specimen extraction. In 1998, Gagner published a series of needlescopic cholecystectomies where a gastrotomy was performed, and the gallbladder was placed into the stomach. After laparoscopic suturing of the stomach, the gallbladder was extracted orally with an endoscope [38]. Although the work was published and successful, local criticism prompted abandonment of the idea.

Kalloo et al. published their landmark paper on transgastric peritoneoscopy in 2004, and this immediately opened the eyes of many to the potential toward intra-abdominal surgical procedures via the natural orifice, in this case, the mouth. Using a swine model, the authors created a gastrotomy using a needle knife and passed a guidewire across the opening. The gastric wall was dilated with a balloon or enlarged with a pull-type sphincterotome. The endoscope was passed into the abdominal cavity, and a liver biopsy was performed. The gastrotomy was closed using endoscopic clips. They performed 12 nonsurvival cases and later 5 survival cases [1]. This series demonstrated that deliberate perforation and subsequent closure of the gastrointestinal tract, with a minor procedure performed through the opening, was safe and repeatable. This study began to fuel the imaginations of both gastroenterologists and gastrointestinal and endoscopic surgeons about the possibilities of incisionless surgery in the abdominal cavity.

Also in 2004, a surgeon and a gastroenterologist team in Hyderabad, India, performed a human NOTES™ case. Rao and Reddy had a patient with severe burn scars on the abdominal wall who presented with acute appendicitis, so they chose a transgastric route to the abdomen [39]. Although this work has never been formally published, they presented a video of this case at both the Society of American Gastrointestinal and Endoscopic Surgeons and Digestive Disease Week annual meetings in 2005. In the operation, they used an endoscope to transit the stomach and used bipolar cautery via the endoscope to divide the mesoappendix. An endoscopic loop was utilized to ligate the appendiceal stump, and a hot snare was used to divide the appendix. Using an overtube, they withdrew the specimen through the mouth. They later reported seven successful cases using this approach in 2010 [40].

In 2005, Kalloo’s group followed their initial work with a report detailing the transgastric ligation of the fallopian tubes in a swine survival study [41]. Six pigs underwent unilateral tubal ligation using endoloops, with the opposite side left intact as a control. Necropsy at two weeks revealed all ligations to be successful both radiographically (hysterosalpingogram) and histologically. There was no evidence of infection or other complications. Also in 2005, Kantsevoy and colleagues performed endoscopic gastrojejunostomy in two pigs. They utilized a prototype suturing device dubbed the “Eagle Claw” to secure a loop of jejunum to the gastrotomy site. Midway through the two-week survival period, both contrast and endoscopic examination revealed patent anastomoses with no evidence of leakage. At the two-week necropsy, there were no signs of infection, abscess, leakage, or adhesions [42]. Park and colleagues in Sweden published their swine series of nonsurvival and survival transgastric cholecystectomies in 2005 [43]. They utilized two side-by-side endoscopes, and all survival cases were successful. The gastrotomy site was closed with an endoscopic suturing technique, which they also used to successfully perform three cholecystogastrostomies. Importantly, they described the concept of utilizing a laparoscopic instrument to facilitate the procedure, which they would later refer to as “hybrid NOTES™”—a hybrid of laparoscopic and NOTES™ techniques. In 2005–2006, Thompson et al. in Boston published two reports using a survival swine model that included transgastric peritoneal explorations, oophorectomy and partial hysterectomy [44, 45]. Endoscopic clips were used for gastric closure. All cases in both studies were successful and without complications. In 2006, Gostout et al. at the Mayo clinic developed a model for appendicitis, creating inflammation of the uterine horn with an injection, followed by endoscopic transgastric resection two days later with a second procedure. This report is also important because it described gastric closure using T-tags rather than endoscopic clips [46]. In 2006, the “Apollo Group” performed transgastric splenectomy in a nonsurvival swine model. The splenic vessels were ligated with endoscopic loops and a single endoscopic clip; the vessels were divided with an endoscopic polypectomy snare. The gastrotomy was enlarged for specimen removal with a sphincterotome and closed with endoscopic clips [47].

Transgastric work on the biliary tree, mostly looking at cholecystectomy, began in the laboratory setting in 2007. These early experiments focused on feasibility and device development, recognizing the need for a flexible instrument platform that could be “rigidized.” [48–54]. The first human cases of transgastric cholecystectomy were reported by Auyang et al. in 2009 [55]. Four transgastric cholecystectomies were completed via a hybrid approach—the cystic duct and artery were ligated with a laparoscopic clip applier. They noted the difficulty of performing the entire case in a retroflexed position, as has been noted by others.

Our group reported initial experience with transoral, transgastric pancreatic pseudocystgastrostomy in 2008 [56]. Our initial patient was a critically ill man with a large infected pancreatic pseudocyst, who was hemodynamically unstable. Two double-pigtail stents had previously been placed endoscopically into the infected cyst, but due to hemorrhage and the presence of debris, endoscopic drainage had failed. We removed the stents, dilated the tract with an endoscopic balloon dilator, and passed a flexible, transoral linear stapler through the opening into the cyst. Firing the stapler created a stapled pseudocystgastrostomy. Further details on this technique are discussed elsewhere in this text.

Transgastric peritoneoscopy was reported by Hazey et al. in 2008 in ten patients [57]. In this pilot series, patients that had a pancreatic mass and were to undergo diagnostic staging laparoscopy prior to potential pancreaticoduodenectomy underwent both laparoscopy and transgastric endoscopic peritoneoscopy. For patients who went on to undergo pancreaticoduodenectomy, the gastrotomy site was resected. For those were not resectable, the gastrotomy site was used for the palliative gastrojejunostomy. The findings at laparoscopy and endoscopic peritoneoscopy were in agreement in 9 of 10 patients, leading the authors to conclude that the approach was safe.

While Ponsky was working on endoscopic surgery of the upper GI tract in the early 1980s, Buess in Germany began work on the lower GI tract with a technique he coined transanal endoscopic microsurgery (TEM) [58]. In 1985, he reported twelve rectal operations with surgical suturing utilizing an operating endoscope [59]. He continued developing the technique, and over the next decade more reports by him and by others emerged. While there were scattered case reports of colectomy via a transanal approach dating back to the 1950s, its use aside from abdominoperineal resection was not popularized until after the development of the laparoscopic approach to colon surgery in the 1990s. In the early 1990s, Franklin in San Antonio began using a transanal approach for specimen extraction after laparoscopic colectomy [60].