The first surgery performed under general anesthesia was in Boston in 1846. Instead of operating on a patient intoxicated with alcohol, the most used “anesthetic” at that time, the introduction of ether resulted in a calm, pain-free patient. The introduction of ether anesthesia was a sensation to say the least, and a major medical breakthrough that allowed for the development of modern surgery [1].

However, it only took 2 years until the first anesthesia death occurred. An unfortunate young woman died from suspected aspiration after a toenail extraction in Newcastle, UK. The operation had gone well, but afterwards the patient was not fully responsive and was given water and brandy by her doctor. She never recovered and very quickly died. Upon autopsy a full stomach was found as well as congested lungs and although the exact cause of death was never established, this was believed to be the result of aspiration [2]. The first certain case of death from aspiration was reported in 1862 [1]. Still, Lord Lister proposed in 1883 that while solids should be avoided before surgery, he proposed a cup of beef tea 2 h before the administration of chloroform [3]. A report on aspirations in women undergoing caesarean section in the USA in the 1940s also recommended a prolonged fasting period to reduce the risk [4]. In this report, none of the patients died. Nevertheless, based on no scientific testing, it was proposed in a textbook in 1964 for the first time that all patients should be fasting from midnight to the day of elective surgery to secure an empty stomach. Since then subsequent textbooks in anesthesia and surgery have stated the same rule. This is probably the best known medical “rule” in the world today, and it is still widely applied despite ample science showing that it has no data in its support.

Eventually, investigators started to challenge these routines. The pioneering work came from Canada and led by Dr. Roger Maltby [5] showing that intake of 150 ml of water fluids 2 h before surgery instead of overnight fasting actually resulted in a lower volume of fluids in the stomach at the time of surgery. Several follow-up randomized trials confirmed that intake of different types of clear fluids resulted in similar or lower gastric volumes at the time of anesthesia compared with those found after an overnight fast. Other large-scale studies of national data reported that aspirations were very rare events in elective surgery [6]. The majority of the aspirations occurred in emergency patients and often during nighttime. Fatal outcomes following aspirations were mainly found in patients with severe comorbidities. It was also found that fasting overnight was no a guarantee that the stomach was empty the next morning. Other factors apart from the period of fasting determine the volume in the stomach at any given time, such as gastric motility, fluid balance, and the type of food consumed.

While the rule of nothing to eat or drink after midnight is simple and potentially easy to follow, it also results in some of the most disturbing and common complaints that the patients have before surgery. Thirst ranks amongst the most common complaints alongside hunger, anxiety, and difficulty to sleep [7], and thirst (and hunger to some extent) can be reduced by allowing free intake of clear fluids. Another complaint is headache from lack of caffeine in coffee drinkers, which can also be overcome by a simple change of practice.

With the growing evidence that allowing oral intake of fluids up until 2–3 h before an operation was safe and had patient benefits in terms of well-being, gradually national guidelines began to change. Guidelines from Canada and Norway were the first to change [8, 9] followed by several European countries and the USA. The recent European Society of Anesthesiology guidelines are also consistent with these recommendations [10] (Table 4.1).

The idea behind the addition of carbohydrates to the oral drink stems from animal studies of severe stress such as near-fatal hemorrhage or endotoxemia. These studies showed that even a short period of fasting depleting or reducing liver glycogen before such stress was associated with more catabolic metabolic reactions and in its extreme also mortality (for review see [11]). These observations led to the idea that the metabolic state of the patient—fed or overnight fasted—may impact the metabolic response to elective surgery and potentially postoperative recovery.

The first studies addressing this in patients were done using a highly concentrated i.v. glucose infusion (20 % glucose infused at 5 mg/kg/min) given in a large vein overnight [12, 13]. The studies revealed a marked difference in metabolic response to surgery, with the infusion of a high load of glucose resulting in less protein loss and a 50 % reduction in postoperative insulin resistance. Because of the high osmolality of the solution however some patients suffered from irritation and even some mild pain near the infusion site, despite the use of a large peripheral vein. To simplify the treatment, a carbohydrate-rich drink was developed, tailored for preoperative use, and further studies on its potential impact were performed.

The main objective of preoperative carbohydrate treatment is to change the overnight fasted state to a fed state through the activation of insulin to the levels seen after a normal meal (about 5–6 times basal fasting levels). In addition, the administered carbohydrates should ensure that glycogen stores were filled. At the same time it was necessary that the drink would be emptied quickly enough to be safe for use in clinical practice within the then newly formed modernized fasting guidelines. This was achieved by using complex carbohydrates as the main source of the carbohydrates, allowing the drink to be hypo-osmolar, which had been shown to be important for faster gastric emptying. The drink that was developed and tested contains 12.5 % carbohydrates and has an osmolality of approximately 265 mosm/kg. Most trials used an evening dose of 800 ml (100 g of carbohydrate) the night before surgery and a morning dose of 400 ml (50 g) given approximately 2 h before the induction of anesthesia. Intake of the morning dose of the carbohydrate drink was shown to empty from the stomach in 90 min, only slightly slower than a similar amount of water [14], while evoking the desired insulin response.