Carl von Liebermeister

Liebermeister studied medicine at the Universities of Bonn, Würzburg and Greifswald. At Würzburg he attended lectures by Rudolf Virchow, and he later worked with Felix von Niemeyer, who became an important mentor. Liebermeister received his medical degree at Greifswald in 1856. He moved with Niemeyer from Greifswald to Tübingen in 1862 and qualified there as a university lecturer after publicly defending 25 theses.^[1][4]. In 1865, Liebermeister became professor of internal medicine at the University of Basel, where the hospital became a centre for scientific medicine.

While still a student, Liebermeister became interested in body-temperature regulation, and went on to investigate the physiology of fever and the regulation of body temperature at a time when characteristic febrile courses were being associated with particular infectious diseases. He regarded fever as a regulated elevation of body temperature and developed biophysical and pharmacological antipyretic treatments, especially for typhoid fever. His work included the use of cold-water treatment and his 1875 Handbuch der Pathologie und Therapie des Fiebers (“Handbook of the Pathology and Therapy of Fever”).

He was also the author of a comprehensive work on cholera called Cholera Asiatica und Cholera Nostras,^[5] a treatise that was included in Hermann Nothnagel's Specielle Pathologie und Therapie.

Liebermeister's rule

Liebermeister's rule describes the approximate relation between body temperature and pulse rate in febrile adults. It conventionally holds that a rise of 1 °C in body temperature is associated with an increase of approximately eight beats per minute in pulse rate. The rule is an empirical clinical guide, rather than a universal physiological constant: the response varies with age, resting pulse rate, medication, cardiac disease, and the cause of fever.^[6]

A pulse rate lower than expected for the degree of fever is termed relative bradycardia, or pulse–temperature dissociation, and is also known as Faget sign. It may occur in several infectious and non-infectious conditions, but is not specific to any one diagnosis and must be interpreted in its clinical context.^[7]

Historical analysis has noted that Liebermeister’s own observations yielded an average increase of about 4.6 beats per minute per degree Celsius, rather than the eight beats commonly attributed to the rule. The latter figure should therefore be treated as a later clinical convention, not as an exact result established by Liebermeister’s original data.^[8]

In 1877, Liebermeister published Über Wahrscheinlichkeitsrechnung in Anwendung auf therapeutische Statistik (“On probability theory applied to therapeutic statistics”), which was remarkably modern in its approach. He rejected the use of fixed probability thresholds to classify therapeutic findings as conclusive or inconclusive, arguing instead that each study should be assigned its particular evidential value. He also put forward a sophisticated Bayesian analysis based on the hypergeometric distribution with a flat prior that could be used to convert trial outcomes into (posterior) probabilities that a treatment was effective. Liebermeister's use of the hypergeometric distribution also freed the analysis of clinical trials from their reliance on the Central Limit Theorem, which underpins the (asymptotic) validity of the Normal distribution as sample sizes become infinitely large. This was a major practical advantage that had eluded statisticians up until this point, allowing reliable inferences to be made from small trials of a dozen or so patients. Liebermeister also stressed that statistical analysis could assess the role of chance but could not itself establish the clinical cause of an observed difference; comparability of treatment groups remained necessary.^[2][3]

Contemporary reception

Liebermeister’s method attracted criticism from contemporaries, including Julius Hirschberg, Johannes Korteweg and Friedrich Martius. Their objections centred on the applicability of mathematical models to clinical data, the comparability of patient groups, prior medical knowledge, and the risk that numerical results could convey unwarranted certainty. Despite containing several fundamental misunderstandings of the underlying mathematics, Martius's critique marked the end of public discussion of Liebermeister's work. It has since been recognised as a major achievement in medical statistics, with the use of the hypergeometric distributions pre-dating Fisher's celebrated Exact Test by half a century ^[9]

He returned to Tübingen in 1871 to succeed Niemeyer as professor and head of the university’s medical hospital, a post he held until his death. He promoted and designed a new hospital building, and reportedly declined professorial appointments at Bonn, Freiburg and Leipzig. As physician to King Karl of Württemberg’s royal household, he was ennobled and became known as Carl von Liebermeister.^[1]```. In 1889, Liebermeister developed signs of serious kidney disease. He died in Tübingen on 24 November 1901, aged 68; a later biographical account identifies the cause as a renal tumour.^[1][10]

• "Liebermeister's grooves": Developmental grooves on the surface of the liver.