Blog | Monday, March 12, 2018

The fusobacterium story as of 2018

Over the past 15 years I have developed an obsession to understand Fusobacterium necrophorum both as a cause of adolescent/young adult pharyngitis and the development of suppurative complications. Most readers know that I started studying Group A strep pharyngitis in 1980 (first publication in 1981), and wrote multiple articles about strep pharyngitis in the 80s and 90s. So this story will start with the misunderstandings of the score that now bears my name.

First, I must recognize the many co-investigators involved in my career. Most research involves teams, and our studies are no exception.

In 1980 I was a newly minted attending physician supervising residents in the “non-acute ER”. A resident asked me a question about a patient with a sore throat. In a rare burst of humility, I admitted that I had no idea whether we should give antibiotics. We defaulted to antibiotics. I then decided to learn more.

Some context will help the story. Rapid tests were not yet invented, nor were cell phones. Working with a financially challenged population, almost half of our patients had no home phone. We had to make a decision in the ER, because we could not wait for culture results. We wanted to treat Group A strep to prevent acute rheumatic fever.

We started to collect data to see if we could predict strep pharyngitis from clinical signs and symptoms. Our patients were 15 years and older. The majority of sore throat visits occurred in patients between 15 and 30.

After running throat cultures and collecting data, I went to the statistics department for help in data analysis. That was a rookie move, I should have talked to them first. A great biostatistician recommended a new SAS procedure, PROC LOGIST. I had no idea what he was saying. He printed out the directions and explanation. I spent 3 months learning how to do SAS and trying to understand what logistic regression was.

We looked at over 10 signs and symptoms and were able to develop a logistic regression equation that included 4 factors (now well known): history of a fever, lack of cough, tonsillar exudates, and swollen tender cervical nodes. Amazingly this model had staying power. Multiple studies validated its predictions. Dr. Mark Ebell wrote a 2000 JAMA Rational Clinical Exam article that first called it the Centor score.

Over these first 20 years, many “experts” have missed the point of the score. We did not develop the score to diagnose Group A strep pharyngitis in adolescents and young adults, rather we showed (as did many subsequent studies) that as the number of these 4 factors increased in patients with acute sore throats, the more likely that the culture would grow group A strep. We never claimed or suggested that the model would work in children (pre-adolescents) because we did not develop it in children. We felt that the model was best at identifying patients with very low risk of Group A strep infection.

We subsequently published data showing the Group C/G pharyngitis looked much like Group A strep pharyngitis.

When I moved from MCV to UAB in 1993, I stopped doing sore throat research. I thought I was done with that chapter of my scholarly activity. Two things happened that drew me back to adolescent/young adult pharyngitis. First, ACP published a guideline that recommended using the Centor score to first identify patients who needed neither antibiotics nor testing, then recommend testing for scores of 2, and either test or treat empirically. The Infectious Disease Society of America agreed with no testing or antibiotics for the 0 and 1 patients, but rapid testing only for the 2, 3, and 4 patients. Members of that guideline publicly castigated (in an editorial) ACP for recommending that empiric antibiotics based on scores of 3 or 4 was rational.

As I read this controversy I became incensed because I thought that the IDSA did not really understand primary care practice. So I entered the fray with several editorials.

I had always wondered what caused the 3s and 4s that were not Group A strep or Group C/G strep. The clinical response seemed more likely to be secondary to a bacterial infection than a viral infection. But what other bacteria might be causing these patients' sore throats.

Then I started this blog in 2002. As you might imagine, articles about sore throats attracted my attention. An article in the BBC news health section stimulated this post, "Some sore throats are VERY serious".

The BBC article, which reported an increase in the Lemierre Syndrome in England, stimulated a new interest, Fusobacterium necrophorum. Over the next several years, I would read about this bacteria and this important suppurative complication.

I subscribed to PubMed and searched for Fusobacterium and pharyngitis. By 2007, four separate articles, two from Great Britain and two from Denmark, documented Fusobacterium pharyngitis. Two studies used anaerobic culture and two used PCR identification. Unfortunately, these studies had no clinical information other than a presentation for a sore throat. One study showed clearly that they diagnosed Fusobacterium pharyngitis much more often in adolescents and young adults than in childhood.

In 2008, a Danish article made me question strongly all the previous assumptions about sore throats. This article documented the Lemierre Syndrome and gave some strong prospective epidemiologic information. This study followed a previous study that suggested an incidence of 1 per million. However, in this study they estimated 14.4 per million in the 15-30 age group. This rare disease now did not seem quite so rare.

This information caused me to think about the risk to adolescents and young adults from Fusobacterium pharyngitis. Much thought led to this perspective published in the Annals of Internal Medicine, “Expand the Pharyngitis Paradigm for Adolescents and Young Adults".

Current guidelines and review articles emphasize that clinicians should consider group A !-hemolytic streptococcus in the diagnosis and management of patients with acute pharyngitis. Recent data suggest that in adolescents and young adults (persons aged 15 to 24 years), Fusobacterium necrophorum causes endemic pharyngitis at a rate similar to that of group A !-hemolytic streptococcus. On the basis of published epidemiologic data, F. necrophorum is estimated to cause the Lemierre syndrome—a life-threatening suppurative complication—at a higher incidence than that at which group A streptococcus causes acute rheumatic fever. Moreover, these estimates suggest greater morbidity and mortality from the Lemierre syndrome. The diagnostic paradigm for adolescent pharyngitis should therefore be expanded to consider F. necrophorum in addition to group A streptococcus. Expanding the pharyngitis paradigm will have several important implications. Further epidemiologic research is needed on both F. necrophorum pharyngitis (especially clinical presentation) and the Lemierre syndrome. Clinicians need reliable diagnostic techniques for F. necrophorum pharyngitis. In the meantime, adolescents and young adults who develop bacteremic symptoms should be aggressively treated with antibiotics for F. necrophorum infection. Physicians should avoid macrolides if they choose to treat streptococcus-negative pharyngitis empirically. Finally, pediatricians, internists, family physicians, and emergency department physicians should know the red flags for adolescent and young adult pharyngitis: worsening symptoms or neck swelling (especially unilateral neck swelling). Adolescent and young adult pharyngitis is more complicated than previously considered. Ann Intern Med. 2009;151:812-815.

While many readers agreed with this perspective, some experts argued that we should not change the paradigm because we did not have sufficient evidence. Does Fusobacterium really cause pharyngitis? Would antibiotics prevent Lemierre Syndrome? Antibiotics have significant problems and we have an antibiotic resistance problem.

Working with colleagues at UAB and a medical student who needed a research project, we developed our own PCR for Fusobacterium necrophorum, the cause of at least 80% of Lemierre Syndrome. Our excellent university health center collected throat swabs for a year from students presenting with sore throats. The results of that study convinced many, but the naysayers still argued against trying to treat empirically. The Clinical Presentation of Fusobacterium-Positive and Streptococcal-Positive Pharyngitis in a University Health Clinic: A Cross-sectional Study

Background: Pharyngitis guidelines focus solely on group A ?-hemolytic streptococcal infection. European data suggest that in patients aged 15 to 30 years, Fusobacterium necrophorum causes at least 10% of cases of pharyngitis; however, few U.S. data exist.
Objective: To estimate the prevalence of F. necrophorum; Mycoplasma pneumoniae; and group A and C/G ?-hemolytic streptococcal pharyngitis and to determine whether F. necrophorum pharyngitis clinically resembles group A ?-hemolytic streptococcal pharyngitis.
Design: Cross-sectional.
Setting: University student health clinic.
Patients: 312 students aged 15 to 30 years presenting to a student health clinic with an acute sore throat and 180 asymptomatic students.
Measurements:Polymerase chain reaction testing from throat swabs to detect 4 species of bacteria and signs and symptoms used to calculate the Centor score.
Results: Fusobacterium necrophorum was detected in 20.5% of patients and 9.4% of asymptomatic students. Group A ?-hemolytic streptococcus was detected in 10.3% of patients and 1.1% of asymptomatic students. Group C/G ?-hemolytic streptococcus was detected in 9.0% of patients and 3.9% of asymptomatic students. Mycoplasma pneumoniae was detected in 1.9% of patients and 0 asymptomatic students. Infection rates with F. necrophorum, group A streptococcus, and group C/G streptococcus increased with higher Centor scores (P<0.001).
Limitations: The study focused on a limited age group and took place at a single institution. Asymptomatic students—rather than seasonal control participants—and a convenience sample were used.
Conclusion: Fusobacterium necrophorum–positive pharyngitis occurs more frequently than group A β-hemolytic streptococcal–positive pharyngitis in a student population, and F. necrophorum–positive pharyngitis clinically resembles streptococcal pharyngitis.

Since Fusobacterium necrophorum recovery increased as the Centor score increased we argued that we had sufficient circumstantial evidence that this organism explained many of the 3s and 4s and that the score really reflected bacterial pharyngitis. Our subsequent recently published paper on the pharyngitis microbiome strongly supports our contentions.

So where are we in 2018. We have increasing information supporting Fusobacterium necrophorum as an important cause of tonsillitis/pharyngitis in the 15-30 age group. It does not appear to be very important in pre-adolescents. We estimate that 1 in 400 Fusobacterium pharyngitis patients will develop the Lemierre Syndrome if they do not receive antibiotics. Another 2-3% will likely develop peritonsillar abscess (as this age group has the highest incidence of peritonsillar abscess and Fusobacterium necrophorum is now the most common organism found in these abscesses.

The Lemierre Syndrome is a devastating infection and while antibiotics successfully treat most patients, these patients often have a horrendous clinical course including significant ICU stays. In my thinking an ounce of prevention is clearly worth a pound of cure. So I continue to advocate empiric antibiotics for scores of 3 or 4 in the most susceptible age group (15-30) with penicillin as the first choice and clindamycin as the alternative.

Sore throats usually improve over 3-5 days, A worsening sore throat raises a red flag. These patients deserve either oral clindamycin or possibly hospital admission for IV antibiotics. Unilateral neck swelling deserves an ultrasound or CT looking for internal jugular thrombophlebitis. Rigors after a sore throat suggests bacteremia and thus we should admit these patients for blood cultures and IV antibiotics.

Some commercial companies have developed or are developing PCR testing for Fusobacterium necrophorum from throat swabs. Unfortunately, PCR testing is rather expensive (generally more than $100).

So in 2018 I still recommend empiric antibiotics based on the clinical presentation. As we study this infection further, perhaps we will develop better targeted strategies, but until then I believe we have an obligation to our patients (aged 15-30) to try to prevent this devastating complication.

db is the nickname for Robert M. Centor, MD, MACP. db stands both for Dr. Bob and da boss. He is an academic general internist at the University of Alabama School of Medicine, and the former Regional Dean for the Huntsville Regional Medical Campus of UASOM. He still makes inpatient rounds regularly at the Birmingham VA and Huntsville Hospital. His current titles are Professor-Emeritus and Chair-Emeritus of the ACP Board of Regents. This post originally appeared at his blog, db's Medical Rants.