Table of Contents
- Key Points
- Introduction: The Revolutionary Impact of Insulin
- Evolution and Revolutions from Inception: The Insulins
- Standardization: Building a Stable and Reliable Insulin Unit
- Purification and the Quest for a Perfect Insulin
- Human Insulin: A New Era
- The Current Revolution: Insulin Delivery, Present and Future
- The Artificial Pancreas and Continuous Infusion
- Portable Pumps and Routes of Administration
- Implantable Pumps: The Next Frontier
- Clinical Implications and a Patient Case
- Limitations of the Review
- Recommendations for Patients
- Frequently Asked Questions
- Source Information
Key Points
- Insulin, discovered in 1921, transformed diabetes from progressive malnutrition into a manageable chronic disease.
- Standardized insulin units took nearly 40 years to develop; international standards were set from 1925 onward.
- Purified and human insulins reduced impurities and lowered the risk of allergic reactions and antibody formation.
- Portable and implantable pumps, along with the artificial pancreas, improved glucose control but required frequent monitoring.
- In one patient, an implantable pump gave satisfactory glucose control and resolved retinal swelling, improving quality of life.
Introduction: The Revolutionary Impact of Insulin
The discovery of insulin in 1921 by the Canadians McLeod, Banting, and Best was a turning point in medical history. The first patient treated, Leonard Thompson, showed spectacular improvement, leading to rapid introduction of insulin therapy at Toronto General Hospital in January 1922, and soon after, worldwide. Insulin became the first therapeutic agent produced by genetic engineering. It also provided a reference model for research in peptide hormones, protein chemistry, crystallography, molecular genetics, cellular biology, endocrinology, and many other fields.
Insulin has changed diabetes from a disease of progressive malnutrition into a new disease entity: insulin-dependent diabetes. This has allowed clearer understanding of the humoral and vascular consequences of diabetes. Current research focuses on controlling not only blood glucose levels but also vascular degenerative complications, including the roles of abnormal blood glucose, hormonal imbalances, direct effects on the vascular endothelium and platelets, and factors affecting blood clotting and viscosity.
Insulin therapy remains at the forefront of research, and this article reviews how and why. It also assesses the current status of insulin in diabetology and general medicine, and the various means of delivering insulin available at the time and expected in the future.
Evolution and Revolutions from Inception: The Insulins
The production of insulin posed major challenges. Extracting insulin from bovine or porcine pancreas required careful coordination with slaughterhouses and immediate extraction under perfectly controlled conditions. It could not be left to isolated medical or pharmaceutical groups.
Within 2 to 3 years (from 1922 to 1926), 12 laboratories in six or seven countries had started to manufacture insulin. This marked one of the starting points of the modern pharmaceutical industry.
The first manufacturers included:
- 1922 – Connaught Laboratories, Canada
- 1922 – Eli Lilly, USA
- 1923 – Allen & Hanburys, UK
- 1923 – Boots Pure Drug Company, UK
- 1923 – British Drug Houses, UK
- 1923 – Burroughs Wellcome, UK
- 1923 – Commonwealth Serum Laboratories, Australia
- 1923 – Farbwerke Hoechst, West Germany
- 1923 – Nordisk Insulinlaboratorium, Denmark
- 1923 – NV Organon, The Netherlands
- 1924 – ER Squibb & Sons, USA
- 1925 – Novo Industri A/S, Denmark
- 1938 – Swiss Serum and Vaccine Institute, Switzerland
- 1945 – Hormon-Chemie, Munich, West Germany
- 1950 – Weddel Pharmaceuticals, UK
Standardization: Building a Stable and Reliable Insulin Unit
Developing a standardized insulin from pancreatic extracts was extremely difficult. It took nearly 40 years to achieve a stable and reliable insulin unit. The early units used rabbit and mouse bioassays, along with a range of others, including the Toronto unit, physiological unit, clinical unit, H unit, and U unit. Eventually, international standards were established:
- 1925: First International Standard – 8 units/mg of amorphous insulin
- 1935: Second International Standard – 22 units/mg of crystalline insulin (Scott)
- 1950: Third International Standard – 23.9 units/mg of crystalline insulin (WHO)
- 1952: Clinical unit defined as 23.95 units/mg of crystalline insulin (WHO)
- 1958: Fourth International Standard – 24 units/ml of crystalline insulin (WHO)
Simplification of dosage with the development of insulins acting for more than 8 hours was of historic importance. The major inconvenience of two, three, or even four daily injections led to the development of long-acting formulations. In 1936, protamine-zinc insulin was introduced. This worked by forming non-covalent complexes of insulin with other proteins, or preparing slow-release insulin crystals. Zinc helped polymerization in both approaches. Other substances, including polyvinylpyrrolidone and carboxymethylcellulose, were tried but never reached the market.
However, delayed-action preparations are no longer considered ideal. Current work aims to simulate the natural pattern of insulin secretion: a constant baseline activity with a major increase after meals. One promising area involves insulin release controlled by glucose levels. Researchers synthesized a soluble complex of sugar, insulin, and the lectin concanavalin A. Insulin is liberated from this complex as a function of glucose levels in the medium. After testing various oligosaccharides, including maltotriose, mannotriose, and mannotetraose, the rate of mannotetraose-insulin release in vitro appeared compatible with the needs of diabetic patients. This slow-release form would eliminate the need for more complex administration systems such as pumps or a miniaturized artificial pancreas.
Purification and the Quest for a Perfect Insulin
The purification of insulin has undergone continual improvement. First, it was necessary to produce rigorously sterile insulin to prevent the insulin abscesses frequently seen in the era of Leonard Thompson. Then, insulin was chemically purified to remove non-specific foreign proteins.
After only one crystallization of bovine or porcine insulin, proteins present in amounts of less than 0.2% were not detectable. These impurities included pancreatic proteins (a-component), proinsulin, intermediates, insulin dimers, and b-component insulin derivatives, as well as other hormones such as glucagon, pancreatic polypeptide, vasoactive intestinal peptide, somatostatin, and c-component.
Through repeated recrystallizations, molecular sieve, and ion exchange chromatography, preparations could contain 99% insulin. This is called mono- or single-component insulin. These efforts were followed by attempts to produce an immunologically inert preparation—one incapable of stimulating the formation of anti-insulin antibodies.
Human Insulin: A New Era
The recent preparation of human insulin by various advanced techniques provides the only hope for a truly non-immunogenic insulin. Human insulin can be produced in several ways:
- Extraction from human pancreas
- Full chemical synthesis from amino acids
- Semi-synthesis from porcine insulin (enzymatic conversion using a reaction developed by Novo Industri, in which the terminal alanine of the B chain is selectively replaced with threonine)
- Biosynthesis using recombinant DNA technology (by Genentech and Eli Lilly, using bacteria)
Bacterial, semi-synthetic, and pancreatic human insulins gave identical single symmetrical peaks on analysis, suggesting structural equivalence with human insulin, though not conclusively proving they are free of isomers.
The question remains whether human insulin represents a true therapeutic advance. One study (Figure 1 in the original article) compared NPH biosynthetic human insulin with NPH purified pork insulin in eight insulin-dependent diabetic patients. Human NPH insulin was more efficient, with significant differences between 9 and 16 hours after injection (p < 0.05 and p < 0.01 at various time points).
The characteristics of human insulin include more rapid uptake from subcutaneous sites, decreased hepatic glucose production, increased metabolic clearance, and a virtual absence of antigenicity. However, it was still too early in 1983 to say whether these changes present real clinical advantages. Human insulin is definitely useful in the rare cases of cutaneous sensitivity caused by allergy to non-human insulin.
The Current Revolution: Insulin Delivery, Present and Future
The history of insulin administration is long. The number of daily subcutaneous injections varied with the type of insulin, customs, and individual practices. Continuous recording of blood glucose showed that despite numerous improvements, normalization of blood glucose in brittle diabetic patients was very imperfect, with values very different from those seen in normal subjects.
Efforts to develop non-injectable forms of insulin—administered nasally, orally, or rectally—were unsuccessful because of total or partial breakdown of the hormone in the body. These routes require insulin doses 7 to 10 times higher than those given by subcutaneous injection. Pellets and multiple jet injections appear more promising, but studies were only beginning. With jet injection, insulin is given by pulsed diffusion rather than needle injection, with efficacy the same as conventional injection.
The Artificial Pancreas and Continuous Infusion
The artificial pancreas, developed between 1974 and 1977, clearly demonstrated that when intravenous insulin was controlled by circulating glucose levels, fluctuations in blood glucose could be greatly limited. Blood glucose curves for poorly stabilized diabetic patients became superimposable on those for normal individuals. This was shown in a patient treated successively with subcutaneous regular insulin injections and with the artificial pancreas: the artificial pancreas produced much more stable glucose levels throughout the day.
Researchers recognized the need for a miniaturized artificial pancreas providing continuous control of glucose. This dream has not yet been fulfilled. While reliable microprocessor-controlled portable insulin pumps have been developed, there was still no portable and reliable system for rapid and automatic measurement of blood glucose in vivo.
Self-monitoring of blood glucose allowed patients to adjust their own insulin doses. However, repeated subcutaneous injections, even when meticulously given, yielded results far inferior to those of the artificial pancreas.
The history of continuous infusion is shown in the following milestones:
- Closed-loop system (artificial endocrine pancreas):
- 1974: Albisser et al.
- 1974: Pfeiffer et al.
- 1975: Mirouze et al.
- Open-loop system (pumps):
- Short-term intravenous continuous insulin delivery: 1974 – Slama et al. (Type 1 diabetes), 1977 – Genuth and Martin (Type 2 diabetes), 1976 – Deckert et al. (brittle diabetes), 1976 – Hepp et al. (Type 1 diabetes)
- Long-term subcutaneous, venous, or peritoneal routes: 1978 – Pickup et al. (Type 1 diabetes), 1979 – Irsigler et al., 1979 – Mirouze et al. (brittle diabetes)
Diabetic ketoacidosis, brittle diabetes, and diabetic pregnancy were the first disorders treated with these approaches.
Portable Pumps and Routes of Administration
Portable syringe-type infusion pumps allowed rapid development of continuous subcutaneous insulin infusion. This form of treatment resembled traditional subcutaneous injections. Currently, more than ten manufacturers market pumps for insulin delivery. However, non-syringe-driven portable pumps, though available or soon to be available, are few and very expensive, limiting the number of patients who can benefit.
Three routes are possible for continuous insulin infusion: subcutaneous, intravenous, and intraperitoneal. With continuous intraperitoneal and subcutaneous infusion, the level of free insulin obtained is lower than with the intravenous route. Despite several difficulties, the intraperitoneal route was considered preferable.
Intraperitoneal infusion requires a much more elaborate technique than intravenous catheterization or subcutaneous injection. It requires full sterile surgical procedures similar to those used in peritoneal dialysis. In one approach, polyethylene catheters covered with a thin layer of silicon and a Dacron sleeve were used to prevent bacterial infection at the entry point. The catheter was placed with 10–15 cm intraperitoneally, 5 cm in a subcutaneous tunnel, and the external 20 cm attached to the pump, with access around the navel.
The peritoneal route is useful only if the external end of the catheter is not frequently manipulated. Therefore, it requires pumps with long autonomy.
Implantable Pumps: The Next Frontier
Development of implantable pumps is the logical next step. Relatively few patients had benefited from these devices by the time of the review (Table 5 in the original). Several implantable pump designs were under evaluation, including:
- Siemens (PFA1): peristaltic pump, 8.5 x 6.0 x 2.2 cm, 180 g filled, 10 ml reservoir, battery life 1 year, variable flow, no alarms, remote control. First implanted April 1981. Route: intravenous or intraperitoneal.
- Metal Bellows (Infusaid): vapor-powered bellows, 8.6 x 2.4 cm, 225 g, 47 ml reservoir, battery life not applicable (vapor powered), constant flow, no alarms, no remote control. First implanted December 1980. Route: intraperitoneal. More than 20 in function.
- Sandia: peristaltic pump, dimensions unknown, 280 g, 5 ml reservoir, variable flow, no alarms, remote control. First implanted January 1981. Route: intraperitoneal. 3 in function.
- Medtronic: pulsatile solenoid, 7.0 x 2.8 cm, 180 g, 20 ml reservoir, battery life 5 years, variable flow, flow alarm, remote control. Not yet implanted in humans.
- Pacesetter: peristaltic pump, 8.1 x 2.0 cm, 170 g, 7.2 ml reservoir, battery life 10 years, variable flow, flow alarm, remote control. Not yet implanted.
Implantable pumps solve the problem of catheter insertion, but they do not solve the major constraints of portable systems: frequent self-monitoring of blood glucose, adjustment of basal and post-prandial rates several times daily, and filling the insulin reservoir every 2–3 weeks.
Clinical Implications and a Patient Case
The author’s experience with an implantable pump involved a single patient: a 23-year-old woman with brittle diabetes. She had achieved spectacular improvement with 15 months on a portable pump. Then a pump was implanted beneath her abdominal muscles, with the catheter emptying into the peritoneum. The reservoir was filled every 3 weeks. The patient monitored her blood glucose seven times daily and could choose basal rates and meal doses using an external programmer.
Results were remarkable:
- Satisfactory control of blood glucose levels
- Complete cessation of episodes of severe hypoglycaemia or ketoacidosis
- Regression of incipient macular retinal oedema (early swelling in the retina)
- Complete transformation in quality of life
In the author’s experience with continuous infusion treatment, technical problems were common but manageable. There was approximately one pump failure and one catheter problem every two patient-years. Catheter problems included partial exteriorisation by involuntary traction (twice) and external rupture (twice). Repairs were always possible without needing to remove the catheter or stop the infusion. Early obstruction of the catheter was not seen with stabilized forms of insulin, and it was never necessary to remove the catheter before 13–15 months of usage.
Rates of infection and catheter problems were given as:
- Frequency of peritonitis in uraemic diabetic patients treated by home peritoneal dialysis: one per 17 patient-years
- Frequency of infections at the entry site in all the author’s experience: one per two patient-years
Actuarial curves (Figure 3 in the original) showed the frequency of pump failure and catheter survival over time.
Technical Challenges of Insulin Infusion
Stability of concentrated insulin stored at body temperature and subject to constant agitation in ambulatory patients is a major problem. Ordinary insulin aggregates into high molecular weight polymers, yielding a crystalline precipitate, inactivation, yellowing of the solution, and obstruction of the catheter. The optimal anti-precipitant was a mixed polymer of polyethylene glycol and polypropylene glycol. Satisfactory results were also obtained by adding amino acids (glutamic or aspartic acid) or forming insulin sulphates. Stabilization was also attempted using the surfactant Genapol. The insulin reservoir itself must not denature its contents; current preference is for polyethylene or titanium.
Limitations of the Review
This review reflects the state of knowledge in 1982–1983. The author explicitly notes that the majority of continuous infusion treatment regimens had been discontinued after only a few months at that time, making definitive evaluation of these techniques unavailable. Human insulin was newly introduced, and it was too early to know whether its benefits would be clinically significant, apart from preventing rare allergic reactions.
The number of patients using implantable pumps was very small, and many devices were still experimental. The author also notes that the data on non-injectable routes (nasal, oral, rectal) were unsuccessful, and studies of jet injection were only beginning. Thus, patient outcomes today have changed dramatically, but the historical development is accurately captured.
Recommendations for Patients
Although this article is historical, it offers enduring lessons for patients:
- Know your insulin type: Long-acting, intermediate-acting (e.g., NPH), and rapid-acting insulins each have different time courses. In 1983, human NPH insulin showed greater effectiveness than pork NPH between 9 and 16 hours after injection. Today, insulin choices should be based on your individual glucose profiles and lifestyle.
- Self-monitoring is essential: Even with new delivery technology, frequent blood glucose monitoring is needed. The patient described monitored seven times daily. You should discuss a monitoring plan with your healthcare team.
- Injection technique matters: The review emphasizes that meticulous injection technique is inferior to the artificial pancreas, but for most patients, injections or pump therapy are practical. Use sterile, properly stored insulin and rotate injection sites.
- Because human insulin is less antigenic, those with allergies or skin reactions to animal insulins should ask about human insulin or modern analogues.
- Consider the delivery system: Portable pumps and implantable pumps show promise but require commitment to glucose monitoring, refilling, and attention to catheter problems. Discuss pros and cons with your diabetes specialist.
- Treat the whole person: The 23-year-old patient experienced complete transformation in quality of life. Control of glucose not only prevents acute emergencies but also helps protect against eye problems like macular oedema.
Always consult your physician before changing your insulin regimen or using a new delivery system.
Frequently Asked Questions
How did insulin therapy change diabetes care after its discovery?
Before insulin, diabetes led to progressive malnutrition. After insulin was introduced in 1922, the first patient improved dramatically, and insulin therapy spread worldwide. It transformed diabetes into a manageable chronic condition, allowing clearer study of its complications, such as blood vessel and nerve damage, and opened new fields of medical research.
What are the different types of insulin and how do they work?
Insulin types differ by how long they act. Long-acting and intermediate-acting insulins, like NPH, provide baseline activity. Rapid-acting insulins cover meals. In a study of eight patients, human NPH insulin was more effective than pork NPH between 9 and 16 hours after injection. Your doctor chooses types based on your glucose patterns and lifestyle.
What is human insulin and why is it important?
Human insulin is produced to match the insulin your body makes. It can be made from human pancreas, semi-synthesized from pork insulin, or made with bacteria using recombinant DNA. Human insulin is less likely to trigger immune reactions. It is especially helpful for rare cases of skin allergy to animal insulins.
What insulin delivery methods are available?
Insulin can be given by daily injections, portable pumps that deliver continuous subcutaneous infusion, or implantable pumps that release insulin into the abdomen. Jet injection without needles was being studied. Nasal, oral, or rectal forms failed because the hormone broke down, requiring doses 7 to 10 times higher. Your specialist can discuss the most appropriate method.
What is the artificial pancreas?
The artificial pancreas, developed around 1974 to 1977, is a closed-loop system that measures blood glucose and automatically adjusts intravenous insulin. In poorly stabilized patients, it produced glucose curves like those of normal individuals. It showed that continuous glucose control was possible, but a miniaturized portable version was not yet available for everyday use.
What were the results of implantable insulin pumps in early use?
In one patient, a 23-year-old woman with brittle diabetes, an implantable pump delivered insulin into the peritoneum. She monitored glucose seven times daily. Results included satisfactory blood glucose control, no severe hypoglycemia or ketoacidosis, regression of early retinal swelling, and a complete transformation in quality of life. However, the number of patients was very small.
What should patients know about insulin therapy today?
Know your insulin type and its time course. Self-monitoring blood glucose is essential, even with pumps; one patient monitored seven times daily. Use sterile, properly stored insulin and rotate injection sites. If you have allergies to animal insulins, ask about human insulin. Discuss delivery system options and any changes with your healthcare team.
When should a patient with insulin-dependent diabetes seek a second opinion about insulin delivery options?
A second opinion about insulin delivery is worth considering if you are thinking about switching to a pump, need frequent dose adjustments, or have had skin reactions to animal insulins. Portable and implantable pumps require daily blood glucose monitoring, regular reservoir refilling, and attention to catheter problems, which occurred about once every two patient-years in the experience described. Since insulin choices depend on individual glucose profiles and lifestyle, a second opinion can help confirm which delivery method fits your situation. Diagnostic Detectives Network provides independent expert second opinions.
Source Information
Original article title: Insulin Treatment- A Non-Stop Revolution
Publication: Diabetologia (1983) 25:209–221, © Springer-Verlag 1983. This is the 14th Claude Bernard Lecture delivered to the European Association for the Study of Diabetes, Budapest, September 1982.
Note: This patient-friendly article is based on peer-reviewed research and a historical review. It does not provide medical advice. Treatments and technologies described reflect the state of the art in 1983 and may be outdated; consult current medical guidance for up-to-date recommendations.
