SFGR Position Statement

Hello,
Thank you for distributing the updated position statement. I am wondering whether there is any interest in loosening the timelines (or not defining timelines) for specimen collection to document a fourfold rise in IgG. I recognize that acute and convalescent is ideal, but I'm imagining a specimen collected slightly beyond the first two weeks of illness or not quite 2 weeks after the first specimen. Do we care as long as a fourfold rise is documented? Are there any jurisdictions that think not including a timeline would affect their confirmed case counts?

Thanks,
Hanna

I have seen the new proposed case definition and while there are some improvements to the "Confirmed case" definition,  I find the criteria for counting "Probable" cases to be lacking scientific merit, or any detailed explanation that justifies the case definition criteria. 

First of all, by allowing any positive (quantitative IFA for IgG) result of >= 1:128 as laboratory evidence of infection, the scientific criteria for what makes a valid immunological response is ignored, and at the very least no reasons are given to support the 90 day period.  For example, if we count SFR cases as "Probable" cases when this positive result in the ninety days post illness onset, we will be counting a lot of cases in which the patients were tested for RMSF in the first six days post illness onset, but as persons do not even develop a detectable immune response to an SFR until at least 7 to 10 days after illness onset, all of the cases counted in the first six days of this 90 day period will have positive test results that are not even related to the illness that was being tested for (i.e., the patient could only have had an older exposure to an SFR agent for it to have been detectable  in the first six days of illness.  Therefore, I would like to see a medical / scientific / immunological explanation for why patients with positive SFR test results in their first six days of illness should even be counted as a Probable SFR case.

Additionally, based on a number of studies of Spotted Fever Rickettsioses carried out in the southeastern U.S. (e.g., Mixson et al., 2006, Gaines & Operario et al., 2014;  Lee et al., 2014, and others), it is apparent that lone star ticks are  the most common cause of tick bites in the southeastern U.S., and that anywhere from 45 to 60% of the lone star tick population carries an
SFR agent called Rickettsia ambliommii - (recently renamed  R. amblyommatis).  Conversely, the detection of known SFR pathogens such as R. parkeri or R. rickettsii in lone star ticks has been uncommon or rare, so what SFR pathogen is causing all of these counted SFR cases??  And if one discounts the idea that R. amblyommatis is a pathogen, most studies have identified Ehrlichia agents as the most common pathogens  carried by lone star ticks.

Based on my own experience, the American dog tick (Dermacentor variabilis); the tick most often associated with RMSF transmission, is an uncommon cause of human tick bites in the southeastern U.S., but lone star tick bites are very common.  Over the past 28 years, I myself have been bitten by more lone star ticks than I can recall, but have not had an American dog tick bite since 1973.  Furthermore, considering how commonly lone star tick bites affect people in the southeastern U.S.,  it would seem that if R. amblyommii (amblyommatis) was carried by at least 45% of the tick population, and was actually a pathogen, then large numbers of people would become sick after lone star tick bites (that does not appear to happen).   And if  R. amblyommatis is not an SFR pathogen, then what SFR agent is causing all of the illnesses being counted by our "SFR" disease surveillance?   

Additionally,  how does counting all of these highly questionable SFR cases  inform the public and scientific community?  Current entomological evidence shows that the most common cause of tick bites in the southeastern U.S. is the lone star tick, and the most common and serious pathogen carried by lone star ticks is ehrlichiosis.   Unfortunately, most state educational outreach materials focus on the American dog tick  as the carrier of RMSF.  Furthermore, the continual focus on RMSF caused by so many questionable  SFR cases  takes the attention of the public and medical community away from lone star ticks, and from a serious disease like ehrlichiosis. 

The great majority of Rickettsiosis patients are seen by doctors only in the acute stage of illness.   Furthermore, many patients are tested positive for RMSF in their acute stage of illness (when they should not even have a detectable immune response to RMSF), and this likely causes doctors to assume that RMSF is the cause, and overlook, or fail to follow up on, serological  testing for ehrlichiosis. 

Instead, doctors should focus their attention on  diagnosis and testing of their acutely ill patients for ehrlichiosis by use of a multiplex PCR assay that tests for both Ehrlichia chaffeensis and E. ewingii.  And if they provide their ehrlichiosis patients with an appropriate course of doxycycline, in the rare event that one of these patients is actually infected with a pathogenic SFR like RMSF, then the course of doxycycline will have done its job, and most of the Rickettsial illness cases identified by doctors will have been correctly identified as ehrlichiosis.   

   

Based on what I have written above, I recommend that at the very least a "Probable" case of SFR would have to meet the following criteria for Presumptive laboratory evidence.

Presumptive laboratory evidence:      

Patient has serologic evidence of elevated IgG antibody at a titer ≥1:128 reactive with spotted fever group Rickettsia, including R. rickettsii, antigen by IFA in a sample taken no sooner than 10 days (and preferably at least 11 days) post illness onset, and no later than 90 days post illness onset.

If any patient has an IgG  titer of ≥1:128  in the first six days after illness onset that immune response could only be attributable to an old SFR exposure that is not related to the illness being tested for.    Although some patients tested by a quantitative IFA for IgG will begin to develop a detectable positive (≥ 1:64) seven days after illness onset, some do not have a detectable immune response until 10 days post onset, and as the development of the immune response would require some additional days before the titer reached a level of ≥1:128, it would be prudent to wait at least until 10 or 11 days after onset of illness before counting a titer of ≥1:128.     

Most patients suffering from a rickettsial disease appear to visit their doctors within the first week of illness onset, and unfortunately they could not have possibly developed a detectable immune response to their current illness in that short period.  And as patients may maintain a detectable immune response to their SFR exposure for a year or more, counting patients who have a detectable immune response in their first week of illness only reinforces the bad medical diagnostic practices, and the notion among doctors and the public that RMSF is a common disease that is detectable by serological assay in the first week of illness.   If anyone can provide an valid explanation that contradicts what I have said above, I would love to read it.

We are also curious as to why so many health departments in the southeastern states are concerned that a revised SFR case definition would significantly reduce the number of SFR cases they are reporting?  Is there a fear that if insufficient SFR cases are reported, some funding sources for tick-borne disease surveillance will be reduced?  Certainly ELC money for tick-borne diseases would not be reduced, because SFR is not the only tick-borne disease of concern, and in the southeastern states where local transmission of Lyme disease is rare or non-existent and lone star ticks predominate, there is certainly no shortage of ehrlichiosis to report.  Furthermore if we reduce the number of bogus SFR cases being investigated by epidemiologists, they would have more time to reach out to the medical community and educate them about the best laboratory diagnostics to use for identification of  ehrlichiosis cases.

One thing is certain, and that is that most rickettsial disease patients are seen by their doctors in their first week of illness onset, and this is always too early in the course of illness for a patient to have developed a detectable immune response to a rickettsial pathogen (e.g., SFR or ehrlichiosis).  Furthermore, it is evident that most patients never return to their doctor to provide a convalescent serum sample, and as there does not appear to be many non-pathogenic ehrlichial agents in the environment, the probability that ehrlichiosis is identified by a serological assay (quantitative IFA for IgG) in the first week of illness is low.  Yet, due to the abundance of Rickettsia amblyommatis and other presumptively non-pathogenic / low-pathogenic SFR agents in our environment, many patients who are tested for RMSF in their first week of illness have a positive IgG result (most likely caused by an old exposure to Rickettsia amblyommatis from old lone star tick bites).  This is the best explanation for why doctors get so many acute SFR positive test results, and is possibly why so many believe that RMSF is rampant and reflexively test for it. 

However, if entomological survey data was used to educate the medical community that exposure to pathogenic ehrlichial agents is relatively more common than exposure to pathogenic SFR agents, and were educated to  rely on the multiplex PCR assay for Ehrlichia chaffeensis and E. ewingii, entomological (lone star tick) survey data suggests that they would often detect one of these Ehrlichia pathogens in patients that have characteristic rickettsiosis symptoms. 

I have heard some doctors balk at the cost of the multiplex PCR assay for Ehrlichia, but if they were properly using serological assays to test their patients for rickettsial pathogens, they would have to test both the acute and the convalescent serum samples, and the cost of serological testing would be greater than the cost of a single mulitplex PCR assay. 

One other important issue related to the multiplex PCR assay is that the only diagnostic Laboratory that offers the multiplex PCR for Ehrlichia chaffeensis,  E. ewingii, and Anaplasma phagocytophilum  is the Mayo Clinic Lab.  Although Lab Corp offers a multiplex PCR assay for Ehrlichia chaffeensis and A. Phagocytophilum,  their multiplex assay does not test for E. ewingii.  And as most doctors and hospitals use Lab Corp as their main diagnostic lab, it is likely that a lot of cases of E. ewingii are being overlooked or misdiagnosed as SFR.

David N. Gaines

Hi Hanna,
The core group considered scenarios such as what you are describing.  The current case definition from 2010 words this as: "paired serum specimens (one taken in the first week of illness and a second 2-4 weeks later)". We loosened it to read "paired serum specimens (one taken in the first two weeks of illness and a second 2–6 weeks later."  We felt that this loosened the guidance while providing standardization.  The current arboviral case definition states "four-fold or greater change in virus-specific quantitative antibody titers in paired sera" so does not give a specific timeline as you are suggesting.  However, in CDC's reference manual (https://www.cdc.gov/lyme/resources/TickborneDiseases.pdf), they give the timeline we see in the 2010 case definition.  We feel that loosening the guidance as we have allows for some leeway (giving a little more time for sample acquisition), while requiring a timeframe for seroconversion to help capture true cases of SFR. 

Let me know if you have any other thoughts about this.

Thanks,

Abelardo

Hi David,

Thanks for your comments from yesterday.  We will look at the second set of comments you just sent and reply soon to concerns not addressed below.  We agree with many of your comments.  Below is a summary of your first set of concerns and our perspective:

1.  Concern: Disagree with timeline of 90 days for accepting a positive sample.  Comment: CDC currently uses 90 days.  Our case definition of acute within 2 weeks of onset and convalescent 2-6 weeks later, we need to be able to accept specimens for 8 weeks from onset or 60 days.  We are discussing 60 vs 90 days and will get back to you about that.

2.  Concern:  Lone star tick bites are more common and do not transmit R. rickettsii but rather ehrlichia and other rickettsia.  Comment:  The goal of the position statement is to capture as many true cases as possible given the variation of SFGR across the US.  Arizona, for example, does not have established lone star tick populations.  You raise many important questions about ticks that need to be studied.  As an example, this paper https://www.ncbi.nlm.nih.gov/pubmed/29771344 discussed the multiple rickettsia found in dog ticks, adding to the confusion about what rickettiae are responsible for the seroconversions in patients.  Regarding ehrlichia in our endemic states, VA, TN and so on, it may be that it is causing seroconversion as well, but even if we added an exclusion criteria of "and not positive for any other endemic TBDs" or "not positive for ehrlichia" specifically, we would still not be able to distinguish between ehrlichia and SFR.  Even with a PCR positive ehrlichia test, we cannot rule out a SFR infection if there is an antibody response.

3.  Concern: Non-pathogenic R. amblyommii causing seroconversion.  Comment:  There is some evidence that R. amblyommii is not non-pathogenic (https://www.ncbi.nlm.nih.gov/pubmed/27022147;https://www.ncbi.nlm.nih.gov/pubmed/25187639).

4.  Concern:  Seroconversions occurring during acute stage are not an indication of current infection.  Comment:  While most people may develop antibodies 7-10 days after an infection, this may vary on an individual basis and depending on the species, dose, etc.  Some literature regarding seroconversion before 7 days:  https://www.ncbi.nlm.nih.gov/pmc/articles/PMC262206/;https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6311067/.  Further, from a practical perspective, most patients arrive to see the doctor in the acute stage and that is when a sample can be taken.

5.  Concern:  Doctors are not focusing on ehrlichia testing. Comment: We agree that physician education is important.  The position statement serves to guide but even if we include an ehrlichia exclusion as mentioned above, it would not clarify whether there was a SFR infection or not.  We can do PRNTs to discern the most likely etiologic agent in flavivirus infections but we do not have a good test to distinguish SFR vs ehrlichia infections.

We hope that helps address your concerns.  The case definition will never be perfect and correctly identify all cases but it will allow us to standardize recommendations based on our current technology and understanding and to identify national trends.  As better diagnostics are available we can continue to refine the case definition, but until then, we hope the changes we are recommending will both better capture true SFR cases while decreasing work load in high incidence states. 

Let me know if you have any additional concerns.  

Thanks,

Abelardo

Tori

It would be great if we could educate doctors to always collect appropriate samples from suspected SFR (RMSF) cases for molecular testing.   Such samples would include a punch biopsy from a macule, papule or petechia in a rash, or from an eschar; or collection of swab samples or scabs from an eschar, or collection of a whole blood sample from patients with a characteristic RMSF rash in which petechiae were present. 

However, it is likely that a significant number of patients who visit a provider with symptoms compatible with a rickettsial disease do not have such rash signs, and in this case the only recourse for a provider is to do a serological test for RMSF or a multiplex PCR test for Ehrlichia / Anaplasma. 

Given that the patients often do not return to their doctors office to provide convalescent sera after they are feeling better, it is probably more useful for providers to run a multiplex PCR  for Ehrlichia (at least in the southern and eastern U.S. where lone star ticks are common;  note: lone star ticks have recently become established in some areas of the southwestern U.S., but I have not read any studies that indicate whether or not they are carrying Ehrlichia pathogens out there).  There are many other tick species in the western U.S., and as there appears to be a great  variety of spotted fever group rickettsial (SFGR) agents found in different tick species, it is probable that many of the western tick species carry some form of SFGR other than just  R. rickettsii.  However, I do not know of (have not looked for or read) any studies from western states that looked for the presence and prevalence of various SFGR agents or other rickettsial agents Western tick species.  In some areas of the southwestern U.S. there are also problems with endemic typhus, R. typhi which can also cause weak cross-reactive positives (1:64 or 1:128) on the test for RMSF.

In any case, it is likely that we will never see an end to patients with an illness that is compatible with a rickettsial disease, who only see their doctor in the acute stage of illness.  Therefore, the issue of how to interpret serological test results on such single serum samples will never go away.      

My own position is that epidemiologists should be discriminating in how  they interpret these single positive serological test results, and not count cases in patients with positive RMSF test results if the blood has been collected in a period before the patient would/should have developed a detectable immune response.  It is the healthcare provider's job to treat patients when RMSF or another rickettsial disease is suspected, but I do not think that it is an epidemiologist's job to count cases of RMSF or any other rickettsial disease unless there is reasonably good clinical and laboratory evidence of that disease.  

David N. Gaines, PhD

State Public Health Entomologist

Virginia Dept. of Health

Division of Environmental Epidemiology

Richmond, VA 

804-864-8112 – Office

804-221-1028 – Cell

804-864-8131 - Fax 

Hi Tori,

Thanks for your message.  We do have language regarding PCR under confirmatory evidence: "Detection of spotted fever group Rickettsia, including R. rickettsii, DNA in a clinical specimen via amplification of a species-specific target by PCR assay."  This is in addition to serologic, biopsy and isolation evidence.  We have the same issues about not being able to get a lot of confirmed cases.  PCR may help but not widely available commercially so we need the other forms of evidence too.  Hope that answers your question.  Let me know if you have more questions.  

Thanks, 
Abelardo