Just to establish a baseline understanding here, most comatose patients are not actually maintained on intravenous feeds. Parenteral nutrition (i.e., IV feeding) is not as common or preferred as enteral nutrition (i.e., feeding given somewhere along the alimentary tract). There are a few reasons for this, but first let's discuss the development of parenteral nutrition (PN).
There are two types of PN, peripheral (PPN) and total (TPN). You might guess from the existence of "total" parenteral nutrition that PPN does not provide complete nutritional support, and you would be correct. Nutritionally complete intravenous solutions are, by their very nature, very concentrated solutions, which can cause inflammation of veins (phlebitis) and be damaging to surrounding tissue if they leak from the intravenous site (extravasation). PPN is thus not suitable for long term use, but can be used for short periods or as adjunct in patients who are getting nutrition in other ways.
Total parenteral nutrition avoids some of the risks of phlebitis and extravasation of PPN by using an intravenous catheter placed into a central vein. The highly concentrated TPN solution is thus rapidly diluted into the bloodstream by the high volume of blood flowing through large central veins. Central venous catheters carry their own risks, however, chief among them the ability of bacteria and other pathogens to travel down them directly into central circulation triggering dangerous conditions like endocarditis and sepsis (Fonseca et al., 2018).
Any sort of effective parenteral nutrition is therefore reliant upon a host of modern medical techniques and materials. Not only does TPN require placing a central venous catheter under sterile conditions, which is then accessed using aseptic techniques, but it also requires nutritional solutions optimized to provide energy and nutrients in way the body can use without causing harm. So naturally early attempts just injected random liquids into veins willy-nilly, including everything from beer, wine, milk, oil, various electrolyte solutions, and both human and animal blood (Dudrick & Palesty, 2011). Did people die from this? Absolutely! Though it may be some small comfort to know many of these experiments were carried out on animals.
Given the challenges of parenteral nutrition, it should be no surprise that it only became a commonly used technique in the mid-20th century, with a great deal of credit going to Stanley Dudrick at the Children's Hospital of Philadelphia (Nakayama, 2017). Successful implementation was built upon experiments which ramped up in the early 20th Century, but were themselves predicated upon Garvey in the 17th Century finally disproving Galenic ideas about circulation, laying the ground work for a modern understanding of the circulatory system.
Speaking of Classic era physicians, enteral feeding has been the much more common approach throughout history, with Hippocrates attesting to the beneficial effects of "clysters." I don't want to fall into the hackneyed trope of the "the Greeks invented everything" though, as Herodotus reported Egyptian use of medicinal enemas, something which is supported by the Ebers Papyrus. Infusions of barley and honey through the anus, however, were not necessarily directed at nutritional support so much as they were intended as a form of medical treatment in and of themselves. Enteral feeding per rectum was recognized at least as early as the Late Roman Republic though, when a physician named Celsus recommended mixtures of wheat, barley, eggs, milk, and marrow as a last resource to maintain patients (Vassilyadi et al., 2013).
Rectal nutrition enjoyed some measure of prominence in the 19th Century, with the use of rubber tubes inserted a foot or more into the large intestine. One prominent paper at the time claimed successfully maintaining patients for up to 9 months on an infusion of milk, eggs, beef broth, and pancreatic gland (the latter thought to aid in digestion of the mixture). Famously, President James Garfield was kept alive for months following his injury by an assassin's bullet by enemas containing defibrinated blood, beef broth, and whiskey (Cresci & Mellinger, 2006). An 1878 report claimed one woman was kept primarily on rectal feedings for 5 years, with it being her sole source of nutrition for over a year (Bliss, 1882).
Certainly there are some problems with rectal nutrition. As Dr. Bliss put it in the text cited above, "the lower bowel is essentially an execratory magazine, and should be so regarded in dealing with this whole question" (ibid., p. 6). Various alternatives were tried during this time period, including injecting foodstuffs directly into the small intestine, bathing the peritoneum in broths, and even early gastrostomies, but none were as popular or convenient as rectal feedings. But instilling nutrition from the bottom up carried the intrinsic problem that the colon is simply not designed to absorb food that way (Barr et al., 2021). Fortunately, the problem of enteral nutrition "from above" also had centuries of development at this point.
Renaissance physicians record experiments in providing nutrition via tubes inserted via the nose or mouth into the esophagus to instill nutritional mixtures directly into the stomach, with the first such recorded attempt by Capivacceus in 1598 (Vassilyadi et al., 2013). Undoubtedly, there had been prior attempts, but it does seem people in the past were more comfortable shoving tubes up each others butts than down their throats. Upper gastrointestinal access is a bit more tricky, given that it can trigger the gag reflex, may inadvertently result in putting a tube in the lungs, and requires materials able to form a tube strong enough and flexible enough to navigate the oro- or nasopharynx down into the esophagus and into the stomach (or even beyond). Leather tubes were common, though one physician reported using a whale bone wrapped in eel skin, but it was the ready availability of rubber that made this route viable (Cresi & Mellinger, 2006) and are essentially what are used today. A further advancement was made in the 1920s with thinner and more flexible tubes using a weighted end to advance through the stomach into the duodenum, which made tube feedings better tolerated and reduced the risk of reflux and aspiration (Vassilyadi et al., 2013).
Comatose patients prior to the advent of modern medicine faced a number of challenges. Having to assume the basic cares of every day life for an unresponsive human being is a herculean task. Even assuming the person is breathing on their own and they are hemodynamically stable, they still require near constant care to ensure basic hygiene, avoid bed sores, prevent contractures, and yes, get proper nutrition. While nutritional support may seem to be a simple concept (put food where food goes!), the actual implementation faces challenges of technique, materials, and composition of the solutions used (a topic I've elided over here).
Parenteral feeding carries risks of serious infections, and peripheral approaches are both unable to provide proper total nutrition while also risking damage to blood vessels and surrounding tissue. Enteral feeding is more analogous to the body's natural processes, but getting the right mix of nutrients into the right part of the body can actually be quite difficult. The use of the rectal approach combined both a very basic understanding that the colon is involved with nutrition with a readily available access point to the alimentary canal. It was only superseded by upper GI access when materials to overcome the challenges intrinsic to that approach were commonly on hand.
Nowadays, long term TPN is typically reserved for patients who lack the anatomical or physiological ability to tolerate enteral feedings. Patients who are going to be on long term enteral feeds, however, often use oro- or nasogastric tubes as a bridge to creating a surgical entrance for a feeding tube directly into the stomach or small intestine. Pre-modern people caring for patients unable to eat, however, did not have the wealth of anatomical and physiological knowledge we currently enjoy, let along access to materials and techniques now so commonplace. So they went with what they knew and and the place they could get to, which meant a tube up the butt.
Barr et al 2021 Bottoms Up: A History of Rectal Nutrition From 1870 to 1920. Annals of Surgery Open 2(1).
Bliss 1882 Feeding Per Rectum: As Illustrated by the Case of the Late President Garfield and Others. The Medical Record July 15, 1882. New York.
Cresci & Mellinger 2006 The History of Nonsurgical Enteral Tube Feeding Access. Nutrition in Clinical Practice 21.
Dudrick & Palesty 2011 Historical Highlights of the Development of Total Parenteral Nutrition. Surgical Clinics of North America 91(3), 697-717.
Fonseca et al 2018 The Relationship Between Parenteral Nutrition and Central Line-Associated Bloodstream Infections: 2009-2014. J Parenteral & Enteral Nutrition 42(1), 171-175.
Nakayama 2017 The Development of Total Parenteral Nutrition. American Surgeon 83(1), 36-38.
Vassilyadi et al 2013 Hallmarks in the History of Enteral and Parenteral Nutrition : From Antiquity to the 20th Century. Nutrition in Clinical Practice 28(2), 209-217.