After Koller: From Cocaine to Procaine, Lidocaine and the Sodium Channel

When Carl Koller showed in the autumn of 1884 that a few drops of cocaine solution could make the surface of the eye insensible to pain, he opened a door that surgeons, dentists and chemists rushed through within months. Historians of anaesthesia describe how the news spread around the world in less than a month, and how almost at once doctors began asking a bigger question: if cocaine could numb the cornea, could it numb a nerve, a patch of skin, a whole limb, the lower half of the body? Over the next fifteen years the answers came one after another — the nerve block, infiltration anaesthesia and spinal anaesthesia — and every one of them began with cocaine, the alkaloid that Albert Niemann had isolated from the Andean coca leaf in 1860.

The same years also showed the price of the drug. Patients died of cocaine poisoning, and some of the surgeons who experimented on themselves became addicted. That double story — a remarkable effect tied to a dangerous molecule — drove chemists to look for substitutes, first among the esters (procaine, a cousin of the B-vitamin-like compound PABA, in 1905), then among the amides (lidocaine, which grew out of Swedish work on a compound from barley, in the 1940s), and later bupivacaine and ropivacaine. In the late twentieth and early twenty-first centuries, cell physiology and structural biology finally explained how all of these drugs work: they block the sodium channels that carry the nerve impulse. This page tells that story as history and science only; it gives no dosing or use information.

Table of Contents

  1. Halsted, Hall and the First Nerve Blocks
  2. Schleich and Infiltration Anesthesia
  3. Bier and Spinal Anesthesia
  4. The Price of Cocaine: Toxicity and Addiction
  5. A Benzoic Ester and the Search for Substitutes
  6. Procaine, 1905, a Cousin of PABA
  7. Lidocaine: From Barley to the Operating Room
  8. Bupivacaine, Ropivacaine and Safety
  9. How Local Anesthetics Block a Nerve
  10. Koller’s Name in Anesthesia Today
  11. Key Research Papers
  12. Connections
  13. Featured Videos

1. Halsted, Hall and the First Nerve Blocks

Koller had shown surface anaesthesia: cocaine dropped onto a mucous membrane, such as the conjunctiva and cornea of the eye, numbed that surface. The next step was taken in New York only weeks after the news of Koller’s Heidelberg communication crossed the Atlantic. The surgeon William Stewart Halsted and his colleague Richard John Hall began injecting cocaine solution not into the tissue to be operated on, but next to the nerve that supplied it.

Their best-known experiments, reported in 1885 and reviewed by López-Valverde and colleagues in 2011, were in the mouth. By injecting cocaine near the inferior alveolar nerve, which carries sensation from the lower teeth, and near the nerves serving the upper front teeth, they produced numbness over the whole region that each nerve supplies. This was the principle of the nerve block: interrupt the nerve’s signal somewhere along its path, and everything downstream of that point falls silent. The historians call this the discovery of dental anaesthesia by nerve blocking, and it is the ancestor of the injection a dentist gives today before filling a lower molar.

Grzybowski’s 2008 historical overview of cocaine and the eye describes Halsted and Hall’s first successful block of the inferior dental nerve, followed by Schleich’s infiltration method, as the developments that came after Koller’s finding. The speed is striking: in barely a year, the effect Koller had demonstrated on a frog’s eye had become a working technique for surgery on the jaw.

Self-experiment as method

Much of this early work was done on the experimenters themselves. Halsted, Hall and their students injected one another to map which nerves numbed which areas. Self-experimentation was a common research method in nineteenth-century medicine, and it had serious consequences here, described in section 4.

Back to Table of Contents

2. Schleich and Infiltration Anesthesia

A second technique took a different route. Instead of finding the nerve trunk, the surgeon could flood the tissue itself — skin and the layers beneath it — with a weak solution, so that the fine nerve endings in that area were bathed in the drug. This is infiltration anaesthesia.

Its name is tied to the Berlin surgeon Carl Ludwig Schleich, who in 1892 introduced infiltration anaesthesia using very dilute cocaine solutions. Grzybowski (2008) lists Schleich’s method among the key developments that followed Koller. The important idea was dilution: by spreading a much weaker solution through a larger volume of tissue, Schleich could numb an operating field while exposing the patient to less cocaine overall. In an era when cocaine poisoning was already a known danger, this was as much a safety measure as a technical one.

Infiltration remains one of the basic ways local anaesthetics are used, for example when a small wound is stitched or a skin lesion removed. The drugs have changed — cocaine was replaced long ago, as later sections explain — but the principle Schleich described is the same.

Back to Table of Contents

3. Bier and Spinal Anesthesia

The most dramatic extension came in Germany at the end of the decade. In 1898 the surgeon August Bier, working in Kiel, injected cocaine into the fluid that surrounds the spinal cord (the cerebrospinal fluid, reached through a needle between the vertebrae of the lower back). The nerve roots passing through that fluid were numbed, and with them the lower part of the body. Patients could undergo operations on the legs and lower abdomen while awake.

Bier published the work in 1899 in the Deutsche Zeitschrift für Chirurgie under the title “Versuche über Cocainisirung des Rückenmarkes” (“Experiments on the cocainisation of the spinal cord”). The history of neuraxial anaesthesia by Brill, Gurman and Fisher (2003) dates Bier’s first intrathecal cocaine to 1898 and traces the whole later line of spinal and epidural techniques back to it.

Bier and Hildebrandt

True to the period’s habits, Bier and his assistant August Hildebrandt tested the method on each other, an episode recounted by the Association of Anaesthetists’ Heritage Centre in its account of cocaine and self-experimentation. Spinal anaesthesia went on to become one of the most widely used techniques in surgery and childbirth, although, again, with drugs other than cocaine.

Taken together, these three steps — Halsted and Hall’s nerve block (1884–1885), Schleich’s infiltration (1892) and Bier’s spinal anaesthesia (1898–1899) — laid out within fifteen years most of the main forms of regional anaesthesia used today. All three grew from Koller’s observation on the eye.

Back to Table of Contents

4. The Price of Cocaine: Toxicity and Addiction

The enthusiasm of 1884 and 1885 was soon tempered. Ruetsch, Böni and Borgeat, in their 2001 history “From cocaine to ropivacaine”, describe how the spread of cocaine anaesthesia was followed by reports of serious toxic reactions and deaths among patients. When cocaine is absorbed into the bloodstream in quantity it acts on the brain and the heart; the reports of the period describe agitation, seizures, collapse of the circulation and death. Because early practitioners had no clear idea of safe limits and often used strong solutions in large volumes, these accidents were not rare.

The second cost was addiction. Cocaine is a powerful stimulant of the central nervous system, and the researchers who handled it daily and tried it on themselves were exposed again and again. Grzybowski (2008) records that addiction among those who used the drug was one of the main reasons doctors began looking for safer agents.

Halsted’s addiction

The best-documented case is Halsted himself. Wright and Schachar (2020) record that while performing anaesthesia research early in his career, Halsted became addicted to cocaine and morphine. Hall, too, became addicted, according to the historical accounts of their self-experiments. Halsted went on to build a new multi-tier residency training programme at the Johns Hopkins Hospital, which became the model for surgical and medical residency training in North America; Wright and Schachar analyse how that programme also helped him hide his addiction while he continued to provide patient care and academic training. These facts are part of the published historical record of anaesthesia; they are reported here as that history.

Cocaine’s other effect on blood vessels

Cocaine also tightens small blood vessels. In surgery of the nose and eye this reduced bleeding, which is one reason the drug kept a narrow place in those specialties long after it was abandoned elsewhere (see section 10). For the wider purpose of numbing nerves, however, toxicity and addiction made the search for a replacement urgent.

Back to Table of Contents

5. A Benzoic Ester and the Search for Substitutes

The way out came from chemistry. Since Niemann’s isolation of cocaine in 1860 and Wilhelm Lossen’s further work on its formula in 1865, chemists had been taking the molecule apart. A central finding, summarised by Biscoping and Bachmann-Mennenga (2000), was that cocaine is an ester of benzoic acid: part of the molecule is a benzoic-acid group joined by an ester link to the rest of the alkaloid.

That observation pointed to a strategy. If the numbing effect depended largely on an aromatic acid joined to a nitrogen-containing chain, perhaps simpler molecules built on the same plan — easier to make, cheaper, and less toxic — would numb nerves without cocaine’s effects on the brain and heart. The coca leaf, in other words, provided the template from which a whole family of synthetic drugs was drawn.

The ester generation, 1891–1930

Ruetsch and colleagues list the ester local anaesthetics synthesised between 1891 and 1930, among them tropocaine, eucaine, benzocaine and tetracaine. Biscoping dates tetracaine to 1930. Each was an attempt to keep the numbing power while losing the danger, and each succeeded only in part. Benzocaine, for example, is poorly soluble in water and found its use as a surface anaesthetic rather than an injection. The compound that changed practice most in this generation was procaine.

Back to Table of Contents

6. Procaine, 1905, a Cousin of PABA

The German chemist Alfred Einhorn spent years making and testing cocaine-like esters. Biscoping and Bachmann-Mennenga (2000) date the arrival of his most successful compound, procaine, to 1905. Einhorn and Emil Uhlfelder later described its chemistry in detail in a 1909 paper in Justus Liebigs Annalen der Chemie: the diethylaminoethyl ester of para-aminobenzoic acid, together with related piperidine compounds.

The PABA connection

That chemical name carries a link to nutrition. Para-aminobenzoic acid, usually shortened to PABA, is the same small molecule once counted among the B vitamins (sometimes called vitamin B10), which bacteria use to make folate. Procaine is built from PABA joined to an amino-alcohol chain; when the body breaks procaine down, the ester link is split and PABA is released. The site’s PABA page describes the compound and its history as a former vitamin.

Safer, but not safe

Procaine was far less toxic than cocaine and did not produce cocaine’s euphoria and dependence, and it became the standard local anaesthetic of the early twentieth century. It was not harmless. Jacob and Kovac (2017) describe how a committee of the American Medical Association, reporting in 1924, concluded that procaine too could cause death in large doses — a finding that, in their account, brought clinicians and laboratory scientists together to study local anaesthetic toxicity systematically. The lesson that every local anaesthetic, not only cocaine, has a toxic threshold has stayed central to the field.

Back to Table of Contents

7. Lidocaine: From Barley to the Operating Room

The next great step came from an unexpected natural source: barley. Holmdahl’s 1998 account of the discovery of lidocaine traces it to the biochemist Hans von Euler (Hans von Euler-Chelpin), who, while studying how genes and enzymes were chemically related in chlorophyll-defective mutant strains of barley, isolated gramine, an indole compound. The chemist Holger Erdtman then synthesised a related compound, isogramine, and found that it had weak anaesthetic properties — an echo of the numbing that Niemann and Lossen had noted with cocaine eighty years earlier.

Erdtman and Nils Löfgren made further amino-amide compounds, but, Holmdahl writes, none could compete with the ester anaesthetics derived from para-aminobenzoic acid, such as procaine. Löfgren and Bengt Lundqvist then followed up these studies and found the amide compound lidocaine. Biscoping and Bachmann-Mennenga date Löfgren’s synthesis of lidocaine to 1943 and call it the first amide-type local anaesthetic. In lidocaine the aromatic ring is joined to the rest of the molecule by an amide link instead of an ester link. Amides are more stable, are broken down in the liver rather than in the blood, and do not release PABA.

Gordh’s clinical tests

The anaesthetist Torsten Gordh carried out the clinical testing of the new drug, according to Holmdahl, and published “Xylocain, a new local analgesic” in the journal Anaesthesia in 1949. In Gordh’s clinical tests lidocaine was such a significant advance over procaine that it was introduced for clinical use in the late 1940s, and Holmdahl, writing in 1998, described it as having been the standard local anaesthetic for half a century, with an excellent safety record — while noting that all local anaesthetics are toxic to nerves in high enough doses. It remains one of the most widely used drugs in medicine and dentistry, and is also used to treat certain heart rhythm disturbances.

The story fits the natural-medicine thread that runs through Koller’s whole wing: an Andean leaf gave cocaine, cocaine’s structure gave procaine, and a cereal grain gave the starting point for lidocaine. The site’s barley page covers the grain as food.

Back to Table of Contents

8. Bupivacaine, Ropivacaine and Safety

Once lidocaine had shown that amides worked, chemists produced a series of them with different properties — faster or slower onset, shorter or longer action. Ruetsch and colleagues (2001) describe two that matter most for the safety story.

Bupivacaine

Bupivacaine was synthesised in 1957 and came into clinical use in 1965. Its long duration of action made it valuable for operations and for pain relief lasting many hours. In the following years, however, reports accumulated of bupivacaine causing serious heart toxicity when it reached the circulation in excess, including heart rhythm disturbances and cardiac arrest that were hard to reverse. Ruetsch and colleagues present these cardiotoxicity reports as the stimulus for the next development.

Ropivacaine

Many drug molecules exist in two mirror-image forms, called enantiomers, which can behave differently in the body. Ropivacaine, introduced in 1996, was developed as a pure single enantiomer (the S-form). Ruetsch and colleagues describe it as the latest step in the long effort, begun with cocaine, to separate the numbing effect from toxicity. Biscoping and Bachmann-Mennenga’s review title, “Local anesthetics from ester to isomer”, sums up the arc from cocaine’s benzoic ester to a purified mirror-image amide.

The history from 1884 to 1996 is therefore a single thread: each new compound was measured against the dangers of the one before, and the problem Koller’s generation discovered in the 1880s — toxicity when the drug spreads beyond the nerve — has stayed the central concern of the field.

Back to Table of Contents

9. How Local Anesthetics Block a Nerve

For most of this history no one knew exactly how these drugs worked. The answer came with the understanding of how nerves carry signals. A nerve impulse is a wave of electrical change that travels along the nerve fibre. It depends on tiny protein pores in the nerve membrane, the voltage-gated sodium channels, which open for a fraction of a millisecond and let sodium ions rush into the cell; that inward rush triggers the same process in the next patch of membrane, and so the signal runs along the fibre. The ability to record the current through a single ion channel, developed by Erwin Neher and Bert Sakmann, is described on the site’s Neher and Sakmann page.

Local anaesthetics — cocaine, procaine, lidocaine, bupivacaine and the rest — block these sodium channels. When enough channels in a stretch of nerve are blocked, the impulse cannot pass that point: the pain signal from the tooth, the skin or the eye never reaches the brain, even though the tissue itself is unharmed.

New detail from structural biology

Körner and colleagues’ 2022 review, “Sodium Channels and Local Anesthetics — Old Friends With New Perspectives”, describes how modern methods, including cryo-electron microscopy and computer modelling of the channel, have added to the classic picture. They note that in the 1970s it was suggested that local anaesthetics could enter the channel pore from the fatty (lipid) part of the membrane, a forward-looking idea at the time. Cryo-electron microscopy structures and mutagenesis experiments have since shown that sodium channels do have side openings, called fenestrations, facing the membrane, which are likely the entrance through which these drugs produce their block. The review also covers how the drugs alter the channel’s opening and closing, why the block grows stronger when a nerve fires repeatedly (use-dependent inhibition), and differences between the several sodium-channel subtypes.

The body has several sodium-channel subtypes, in nerves, heart muscle and the brain. Because local anaesthetics act on this whole family of channels, a drug that escapes from the nerve into the circulation in excess can disturb the heart and the brain — the biological basis of the toxicity that worried doctors from the 1880s onward.

The same mechanism links the whole family back to the coca leaf: the property that made Andean chewers’ mouths go numb, and that Koller turned into eye anaesthesia, is sodium-channel block.

Back to Table of Contents

10. Koller’s Name in Anesthesia Today

Cocaine’s narrow place in ophthalmology

Cocaine did not disappear from medicine at once. Altman, Albert and Fournier, in “Cocaine’s use in ophthalmology: our 100-year heritage” (1985), reviewed the drug’s century in eye care, from Koller’s discovery to the limited uses still current in the 1980s. Grzybowski (2007) notes that cocaine’s anaesthetic indications are now strictly limited because of its harms and the risk of addiction. For the routine numbing of the eye, nose and skin, the synthetic agents described above took its place.

The Carl Koller Medal and Lecture

Koller’s name lives on in regional anaesthesia, the specialty that grew from his observation. The European Society of Regional Anaesthesia and Pain Therapy (ESRA) awards a Carl Koller Gold Medal; van Zundert (2010) reported its award to Dag Selander at the society’s 27th annual congress in Genoa in 2008. The society also hosts a Carl Koller Lecture; Neal’s 2011 Carl Koller Lecture on education in regional anaesthesia was published in 2012.

A centennial view

Writing on the hundredth anniversary in 1985, and again in later reviews such as Calatayud and González’s “History of the development and evolution of local anesthesia since the coca leaf” (2003) and Goerig, Bacon and van Zundert’s “Carl Koller, cocaine, and local anesthesia: some less known and forgotten facts” (2012), historians have placed Koller’s 1884 paper at the head of a line that runs through every nerve block, every dental injection and every spinal anaesthetic given since. Koller’s own life after 1884, his move to New York and his later years are told on the Life and Career page.

Back to Table of Contents

Key Research Papers

  1. Koller C. On the use of cocaine for producing anæsthesia on the eye. The Lancet. 1884;124(3197):990-992. DOI: 10.1016/s0140-6736(02)28859-5
  2. López-Valverde A, De Vicente J, Cutando A. The surgeons Halsted and Hall, cocaine and the discovery of dental anaesthesia by nerve blocking. Br Dent J. 2011;211(10):485-7. PubMed PMID: 22116238
  3. Grzybowski A. Cocaine and the eye: a historical overview. Ophthalmologica. 2008;222(5):296-301. PubMed PMID: 18566545
  4. Bier A. Versuche über Cocainisirung des Rückenmarkes. Deutsche Zeitschrift für Chirurgie. 1899;51(3-4):361-369. DOI: 10.1007/bf02792160
  5. Brill S, Gurman GM, Fisher A. A history of neuraxial administration of local analgesics and opioids. Eur J Anaesthesiol. 2003;20(9):682-9. PubMed PMID: 12974588
  6. Wright JR Jr, Schachar NS. Necessity is the mother of invention: William Stewart Halsted’s addiction and its influence on the development of residency training in North America. Can J Surg. 2020;63(1):E13-E19. PubMed PMID: 31944636
  7. Ruetsch YA, Böni T, Borgeat A. From cocaine to ropivacaine: the history of local anesthetic drugs. Curr Top Med Chem. 2001;1(3):175-82. PubMed PMID: 11895133
  8. Biscoping J, Bachmann-Mennenga MB. [Local anesthetics from ester to isomer]. Anasthesiol Intensivmed Notfallmed Schmerzther. 2000;35(5):285-92. PubMed PMID: 10858837
  9. Einhorn A, Uhlfelder E. Ueber den p-Aminobenzoësäurediäthylamino- und -piperidoäthylester. Justus Liebigs Annalen der Chemie. 1909;371(2):131-142. DOI: 10.1002/jlac.19093710203
  10. Jacob JS, Kovac AL. Procaine and Local Anesthetic Toxicity: A Collaboration Between the Clinical and Basic Sciences. Reg Anesth Pain Med. 2017;42(6):760-763. PubMed PMID: 28953509
  11. Holmdahl MH. Xylocain (lidocaine, lignocaine), its discovery and Gordh’s contribution to its clinical use. Acta Anaesthesiol Scand Suppl. 1998;113:8-12. PubMed PMID: 9932112
  12. Gordh T. Xylocain, a new local analgesic. Anaesthesia. 1949;4(1):4-9. PubMed PMID: 18101361
  13. Körner J, Albani S, Sudha Bhagavath Eswaran V, Roehl AB, Rossetti G, Lampert A. Sodium Channels and Local Anesthetics-Old Friends With New Perspectives. Front Pharmacol. 2022;13:837088. PubMed PMID: 35418860
  14. Altman AJ, Albert DM, Fournier GA. Cocaine’s use in ophthalmology: our 100-year heritage. Surv Ophthalmol. 1985;29(4):300-6. PubMed PMID: 3885453
  15. Grzybowski A. [The history of cocaine in medicine and its importance to the discovery of the different forms of anaesthesia]. Klin Oczna. 2007;109(1-3):101-5. PubMed PMID: 17687926
  16. Calatayud J, González A. History of the development and evolution of local anesthesia since the coca leaf. Anesthesiology. 2003;98(6):1503-8. PubMed PMID: 12766665
  17. Goerig M, Bacon D, van Zundert A. Carl Koller, cocaine, and local anesthesia: some less known and forgotten facts. Reg Anesth Pain Med. 2012;37(3):318-24. PubMed PMID: 22531385
  18. van Zundert A. Carl Koller Gold Medal Award to Dag Selander at The 27th Annual Congress of the European Society of Regional Anaesthesia and Pain Therapy; Genoa, Italy; September 24-27, 2008. Reg Anesth Pain Med. 2010;35(1):106-7. PubMed PMID: 20054267
  19. Neal JM. Education in regional anesthesia: caseloads, simulation, journals, and politics: 2011 Carl Koller Lecture. Reg Anesth Pain Med. 2012;37(6):647-51. PubMed PMID: 23086349

PubMed Topic Searches

  1. History of local anesthesia
  2. Halsted, cocaine and the nerve block
  3. Bier and the history of spinal anesthesia
  4. Lidocaine history and Löfgren
  5. Local anesthetics and the sodium channel

Further Reading

Back to Table of Contents

Connections

Back to Table of Contents