Au Conntraire, professeur Pasteur!


The belief that living organisms could come into being spontaneously can be traced back as far as the Ancient Greeks, at least, who thought that the sun's heat formed plants, animals and humans from a primordial terrestrial slime, to which air imparted life, motion and thought.1, 2 This concept was embodied in its most sophisticated form in the 4th Century CE by Aristotle, who held the view that the characteristics of living creatures resulted from a mixture of the four elements of earth, air, fire and water, and the four principles of hot, cold, dry and wet, to which was added pneuma, the animating force or, in Latin, anima, the soul.3 From the time of Aristotle through the Middle Ages to the 17th Century, simple organisms, at a minimum, were certainly thought to generate spontaneously. Johannes Baptista van Helmont, the Flemish physician, alchemist and philosopher, and some say the first biochemist,4 whose life spanned goodly parts of the 16th and 17th Centuries, believed that mice were spawned when old dirty clothes, preferably sweaty ones, were incubated with wheat husks for a few weeks, in the dark and the warmth.5, 6 St. Augustine, Thomas Aquinas, Isaac Newton, René Descartes, William Harvey, Jean-Baptisté Lamarck, and many others, were believers too. But even though Homer knew that larvae appeared in putrefying carcasses only when touched by flies,7 it was not until Francesco Redi published his classic experiments in 1668, that the myth that maggots are bred spontaneously from meat, was exploded.8 Redi, poet and court physician to two grand dukes of Tuscany and a member of the renowned Academia del Cimento, left various types of meat in vessels that were either completely sealed, or left open to the air with or without a cover of the finest Naples gauze. In so doing, he demonstrated that direct contact with flies was necessary for maggots to appear. The fact that Redi showed that larger organisms did not arise by spontaneous generation did not deter the 18th English Catholic priest and naturalist, of Welch origin, John Turberville Needham from concluding, as a result of his carelessly imperfectly conducted experiments, that microscopic organisms — first recognized and referred to as animalcules by the pioneer Dutch microscopist Antony van Leeuwenhoek (1632–1723) — could arise spontaneously, even in boiled mutton gravy sealed in a corked phial.9 Needham, whom Voltaire called a fool,10 was ultimately proved wrong by Lazzaro Spallanzani (1729–1799), the great Italian naturalist, abbot and master of experimentation. In the most exhaustive series of experiments imaginable, on boiled infusions of every kind of seed, grain, bean, pulse, etc. as well as of ova, silkworms, maggots and tadpoles, from which air was reliably excluded or removed, Spallanzani proved beyond doubt, and to Needham's chagrin and aggravation, that spontaneous generation of animalcules was fiction11. Yet in spite of Spallanzani's efforts, the belief in spontaneous generation was upheld for almost a century more until the final blows were dealt by the English physicist John Tyndall in his confrontation with Henry Charlton Bastian12 (partially conducted through the correspondence columns of The Times) and as a result of the famous showdown of science, personality and politics between Louis Pasteur and Félix Archimède Pouchet, the Director of the Natural History Museum in Rouen, France.13

Pouchet's assertion was that life may be created from inert organic matter derived from previously living organisms,14 which he began presenting in 1855 in a series papers to the Academy of Sciences in Paris. Pasteur, a chemist by training, fresh from his success in crystallography and alcoholic fermentation, objected to Pouchet's ideas. In competition with him for the 2,500 francs Alhumbert Prize that was offered by the French Academy of Science to whoever could shed “new light on the question of so-called spontaneous generation”, Pasteur offered his winning essay and described elegant experiments with his famous “swan-like” neck flasks, which showed that sterility is maintained as long as dust and microbes do not have access to the incubation medium.15 While there has been some suggestion of bias by the study section that judged the competition and even complicity, dare one say, on Pasteur's part,16 there is no doubt that Pasteur was right (and Pouchet was wrong), and his later work on Germ Theory of Disease, silkworms, anthrax and rabies clinched his preeminence in the field of contagion and infection. Ironically, the modern hypothesis of the Russian biochemist Aleksandr Ivanovich Oparin (1894–1980) that life on earth may have actually started from simple solutions of organic substances,17 may yet give some small comfort to Pouchet's memory.18 Nonetheless, under ordinary circumstances, spontaneous generation is discredited. Little wonder, then, that the idea that spontaneous infection could occur without injury or penetration in the closed culture medium of ascitic fluid, was treated with skepticism when it was first broadcast 40 years ago.18, 19, 20 The thought that the dark damp warmth of ascites might be an ideal venue for spontaneous generation was not promulgated widely until the independent and almost simultaneous publication of case collections by Kerr et al.18 and Conn,19 despite the appearance of several reports of ascites infection with coliform bacilli, from the waning of the 19th Century20 through the 1950s.22–30Conntradiction of Pasteur was in the air.

My first encounter with the world of Harold O. Conn, the man who more than any other put spontaneous bacterial peritonitis on the hepatological map with his landmark report20 and more than 30 subsequent articles, happened in his absence, when as a new fellow at Yale University, I first attended the legendary Thursday lunchtime pizza meeting of the Liver Service at the West Haven, Connecticut, Veterans' Administration (VA, later Veterans' Affairs) Medical Center. The junior faculty member who had been slated to present his research that day apologized for not showing his first-line studies because “the guy with the funny haircut” (Fig. 1) had taken the Kodachrome™ slides with him on his current lecture tour. Parenthetically it should be noted to those unfamiliar with Dr. Conn's Beatle-like coiffeur, that his signature tonsure is but a mere successor to the Afro-like shock that he sported, before both the hairstyle and the name itself were fashionable. To the author, Harold Conn will always be “Bluff King Hal”, for his good-natured, blunt, frank and plain-spoken style and humor, rather than because of any physiognomical or personality similarity to Henry VIII. A former Johns Hopkins intern, and resident and Chief Resident at the Grace-New Haven Community Hospital at Yale, Conn first became aware of the syndrome that he was to popularize, early on in his appointment as a faculty member of the Yale University School of Medicine, assigned to (and later Chief of) the Liver Disease Service at the West Haven VA. This transpired shortly after he spent 2 years as the first formally-appointed postdoctoral clinical research fellow in hepatic disorders to the renowned Gerald Klatskin. It was early in 1958 that Conn first saw a patient with cirrhosis with what appeared to be a completely new disorder since, at that time, neither he nor Gerald Klatskin were seemingly aware of the aforementioned reports22–30, mostly in foreign tongues, concerning patients with a similar illness. Conn was later to conclude that the terminal illness of Ludwig van Beethoven was probably the first known example of this syndrome,31 from which that titan of music suffered almost 75 years before the first formal clinical description of the affliction was published.23 In addition to being the first patient known by name and fame to have undergone large volume paracenteses repeatedly,32 Beethoven had also suffered immediately beforehand from rigors and severe abdominal pain that heralded the accelerated onset of his ascites.31 The report from Newcastle and Bristol in the United Kingdom19 included patients who had undergone previous paracentesis, some with multi-organism infection and tuberculosis, and others with frankly purulent ascites. In contrast, in each of the 6 episodes in the 5 patients from West Haven, the illness was apparently spontaneous (i.e., without apparent cause) and could not be related to previous procedures. In each case there was a single offending organism in the ascites (E. coli on 4 occasions, and Aeromonas liquefaciens and Enterococcus on 1 each), that was associated with simultaneous bacteremia in 5, and turbid but not frankly purulent ascites with a neutrophil count of less than 15,000 in all.20 All 5 patients succumbed to complications of cirrhosis within 3 months. Conn postulated that the disease occurred because of failure of the cirrhotic liver to filter bacteria that had escaped from the “cirrhotic” intestine into portal blood (rather than from a bloodborne infection elsewhere), leading to infection of the ascites with a secondary phase of systemic bacteremia therefrom.

Figure 1.

Hal Conn in his heyday (no allusion to his crowning glory is intended).

Once spontaneous peritonitis with bacteremia in Laennec's cirrhosis was authenticated as a homogeneous reproducible syndrome20 and given the pithy name of “spontaneous bacterial peritonitis”,21 this clinical entity became widely recognized, including its occurrence in non-alcoholic cirrhosis33 and even in patients without cirrhosis.34–36 The label “Conn's syndrome” might have been applied had it not already been claimed for Harold's elder brother Jerome, a distinguished endocrinologist at the University of Michigan where, incidentally, Harold was a medical student. Jerome W. Conn (1907–1981) first presented the syndrome that bears his name, i.e., primary hyperaldosteronism, in his presidential address at the 27th Annual Meeting of the Central Society for Clinical Research in Chicago, on October 29, 1954.37 Since the term “Neo-Conn” syndrome would probably raise some eyebrows in the current politically-charged climate of the United States, perhaps the choice eponym for the younger brother's discovery could be Conn's syndrome II.

With the passage of time, the number culpable organisms that cause can spontaneous peritonitis has grown from a few common coliforms (E. coli, Klebsiella pneumoniae and Aeromonas liquefaciens), enterococcus and pneumococcus, to over 70 species that are mostly residents of the human gut, encompassing numerous gram-negative bacilli, citrobacter, paracolon and salmonella species, and both gram-negative and gram-positive cocci.38 Spontaneous peritonitis is most uncommonly caused by anaerobes and micro-aerophilic organisms, inherent pathogens (like mycobacteria, neisseriae and dysentery salmonellae), and miscellaneous species like candida, nocardia and vibrio. One might imagine that the literature has been saturated with pathogens but publication is still assured for any report of a case of spontaneous bacterial peritonitis that is caused by an unusual or exotic animalcule with a fanciful name; Corynebacterium xerosis,39Morganella morganii,40Plesiomonas shigelloides,41Weeksella virosa,42Salmonella dublin,43Alcaligenes xylosoxidans44 and Hafnia alvei,45 spring instantly to mind.

Within a few years of the publication of the classical florid presentation of spontaneous bacterial peritonitis,19 it was apparent that patients may show a wider variety of clinical features than was originally appreciated.20 Over the past 40 years the clinical spectrum of the disease has continued to evolve, along with developments in diagnosis, treatment and understanding of pathogenesis, as reviewed very recently by Dr. Garcia-Tsao with succinct erudition.46 Patients do not necessarily suffer all the typical symptoms and signs of peritoneal infection. In some, the disease may present as isolated fever, or exacerbation of ascites (as in Beethoven's case) or with mild nonspecific abdominal pain. Other patients may simply deteriorate, develop renal failure or have isolated encephalopathy.47 Yet others may actually be asymptomatic.48 A necessary condition for spontaneous peritonitis is the presence of ascites, which is usually substantial but on rare occasions it may be difficult to detect clinically.49 Similarly, whereas direct inoculation of hemoculture bottles with ascites increases the yield and speed of obtaining positive results,50 apparent sterility of ascites in the laboratory has been as great as 60% in some recent series.51 Thus, notwithstanding the unarguable desirability of a positive ascites culture (which can even occur in the absence of ascitic neutrocytosis as so-called bacterascites — another variant of spontaneous peritonitis), an ascitic polymorphonuclear neutrophil count greater then 250 mm−3 is the accepted diagnostic parameter in the appropriate clinical setting.52 It remains to be seen whether the use of reagents strips and/or automated blood cell counters will supplant the time-honored artisan practice of manual polymorph counting.53

Contrary to the impression given in the early reports, it happens that bacterial peritonitis does occur spontaneously in cirrhotics in the community,48, 54 and is not solely a consequence of hospitalization, although there may be a different spectrum of offending bacteria in these two settings. Also, additional risk factors have been identified, notably recent gastrointestinal hemorrhage55 and low protein concentrations in the ascites.56 Spontaneous bacterial empyema is another clinical variant of the peritonitis syndrome.57 From the therapeutic perspective, the recent addition of intravenous albumin to the armamentarium of antibiotics has proved to be truly life-saving by mitigating renal impairment58 and reducing mortality59 therefrom. Last, the pathogenesis hypothesized by Conn for the transit of bacteria from the gut lumen to the peritoneal cavity has been validated, with modification and enhancement, by the results of experiments conducted subsequent to his initial report. In patients with cirrhosis, there is intestinal bacterial overgrowth60, increased intestinal permeability,61 small intestinal dysmotility,60, 62 bacterial translocation across the bowel mucosa,63 and impaired bacterial removal by defenses in the lymphatics and the systemic circulation.47, 64 What has yet to be explained convincingly is why translocation of a single organism dominates.65

Before closing it is tempting to mention briefly another spontaneous complication that may threaten the life of the cirrhotic patient with ascites. “The Flood syndrome,” reported in 1961 by Frank B. Flood during his residency in New York City,66 is aptly named. Flood described graphically the outpouring of fluid that occurred in each of 5 patients with alcoholic cirrhosis and prodigious volumes of ascites, when spontaneous perforation of an umbilical hernia led to spontaneous “auto-paracentesis”, with a fatal outcome in each case. The recommended treatment of this rare complication of ascites is to stabilize the patient with intravenous fluids, antibiotics and a sterile occlusive dressing, followed by prompt elective herniorrhaphy and aggressive treatment of the ascites, which may necessitate the placement of a transjugular intrahepatic portosystemic shunt (TIPS).67

Finally, for those who have missed the unmistakable personage of Hal Conn these past few years at national and international liver meetings, be assured that he is well as he approaches the ripening quiet of octogenarianism. His academic attention has been diverted of late from hypertension in the portal system to the pressure within the central nervous system, having successfully been relieved of adult hydrocephalus. He is now undoubtedly an expert on this condition too and will likely soon provide the medical community with a learned review of all that is to be known about “Conn's syndrome III”. With his ready wit, Hal is obviously still capable of spontaneous generation too.


The author gives thanks to Marilyn Barr Conn for giving her permission to reproduce her husband's likeness here, and to Hal Conn for asking her. Margie Myers gives the author carte blanche to thank her for manuscript skills and support.