Tighter Screening and Follow-Up: Keys to Better Syphilis Care With HIV
An Interview With Khalil G. Ghanem, M.D., Ph.D.
Mascolini: Are syphilis rates changing in your patient population with HIV and at risk for HIV?
Ghanem: I'll first give an overview of the US, then focus on Baltimore to give you a sense of how syphilis rates are changing. Within the US for the last 10 years we've seen increasing rates of syphilis in men. In 2010, which is the year with the latest nationwide data, overall syphilis rates went down for the first time in 10 years. But rates of syphilis in men and particularly among men who have sex with men (MSM) continued to go up.
The male-to-female ratio of syphilis was 1-to-1 or 1-to-2 back in 1996, and now it's essentially 7-to-1 male to female, largely because syphilis rates in MSM have increased dramatically. If you look at early syphilis cases -- the cases used to calculate syphilis incidence per year -- almost 70% of new cases in 2010 are among MSM.
Of course, MSM are also at risk for HIV. We don't have population-level data in the US on HIV and syphilis coinfection, but results of studies in different locales indicate that rates of HIV coinfection among people with early syphilis are anywhere between 20% and 70%. In other words about 20% to 70% of patients who have early syphilis are HIV infected.
So syphilis is a huge problem, particularly among MSM with or without HIV.
Mascolini: And what's going on in your cohort?
Ghanem: Baltimore has had very high rates of HIV and syphilis for a long time. Heterosexual men and women had accounted for most syphilis cases in Baltimore until the last 5 to 7 years, when we saw a dramatic shift in the epidemiology of syphilis. Now the vast majority of our cases are occurring among men, particularly young men who have sex with men. And about 30% to 35% of our cases are occurring among HIV-infected men who have sex with men.
Thus Baltimore essentially mirrors the great shift that we've seen in syphilis epidemiology across the country over the last 10 years. For a while Baltimore lagged behind other major cities like New York and San Francisco in terms of this shift. It happened more recently in Baltimore, but the shift to MSM here has also been dramatic.
Mascolini: I know there's a large HIV-positive injection drug-using population in Baltimore. Have you noticed anything about syphilis trends in injection drug users?
Ghanem: Gender is the most distinguishing feature of our syphilis cases -- in other words young MSM and not so much underlying IV drug use. There is a strong association between meeting partners on the Internet and syphilis, and there is a strong association between syphilis and the use of drugs like crystal meth. But the typical IV drug population that we see has not been the main focus of these new syphilis infections.
Mascolini: How often should HIV-positive and at-risk people be tested for syphilis?
Ghanem: I think the CDC has it down pat: The most important thing that we can do as clinicians is to screen our HIV-positive patients to make sure they're not infected, because we have excellent drugs to treat syphilis and thereby prevent its downstream complications. The recommendation that says screen your at-risk and HIV-positive patients for syphilis at least once a year is an excellent recommendation. But here the key words that a lot of clinicians forget are at least. It doesn't mean you should screen for syphilis on a yearly basis for everyone. You should do it at least on a yearly basis for everyone. There are patients I screen for syphilis monthly.
Screening frequency really depends on patient risk factors and behaviors. Based on these factors you can make a decision on how frequently to screen for syphilis. Patients with new sex partners, patients who meet anonymous sex partners on the Internet -- these types of patients need to be screened much more frequently for syphilis than once a year. I don't think anybody would accuse you of screening too frequently. I think most clinicians don't screen their patients nearly enough.
Mascolini: How do you get patients to come in for screening more often than they might for regular HIV checkups?
Ghanem: One thing we can do is remind our patients regularly about syphilis and about the risks associated with syphilis. And remember that patients don't have to be seen by the physician to be screened for any STD. They can just stop by the clinic and get their blood drawn. I've done that with some patients who come for a regular checkup every 3 months but get screened for syphilis every month. They come by in the morning, get their blood drawn, and go to work.
The other option is to send them to the STD clinic for regular screening. I run the STD clinic here in Baltimore, and we see a lot of people who come in and request syphilis testing. We screen them without asking any questions or seeing how often they've been tested for syphilis. So I think the issue of inconvenience can be bypassed because the clinician doesn't necessarily have to see the patient.
I think every 2 to 3 months is usually fairly adequate for someone who has very high risk. And I think yearly is pretty adequate for people who have a low to moderate risk.
Mascolini: Do you get good screening adherence from people you want to test every 2 or 3 months?
Ghanem: No. That's the tough part. If you plan to test someone every 2 to 3 months, you're lucky if you get them twice a year. When you try to test people yearly, you wind up actually testing them every other year. I feel by pushing it -- by aiming for the ideal -- you're much more likely to get a reasonable screening frequency. I push to get very high-risk people tested every 2 to 3 months; I wind up getting them tested maybe once or twice a year. But I can live with that.
Mascolini: Do you have a handle on the syphilis reinfection rate in the people you see with syphilis?
Ghanem: In our population our reinfection rate in the 3 years after initial syphilis treatment is about 30%. It varies depending on the population and the locale, but reinfection rates between 10% and 40% have been documented. Of course someone who gets syphilis once is much more likely to get a new syphilis infection than a person who has never had syphilis. That's why it's really important to follow these patients up, not only to determine their response to therapy but also to make sure they didn't get reinfected.
This is a common theme with almost all sexually transmitted diseases. That's why the CDC recommends rescreening 3 months after treating for gonorrhea, chlamydia, or trichomoniasis. And it makes sense because someone who comes in with syphilis is telling you they have sex in a network that has a high rate of syphilis, and usually networks don't change significantly after a person gets treated. As a result, they run a high risk of getting reinfected if they've already been infected once. It makes perfect sense on the network level, and it makes perfect sense on the biological level.
Mascolini: When should lumbar puncture be done for HIV-positive people with syphilis?
Ghanem: That remains an area of huge controversy for HIV-infected patients with syphilis. There are physicians who feel that all HIV-infected patients should undergo lumbar puncture. The CDC addressed this topic in its 2010 STD treatment guidelines.1
There are three categories of patients in whom lumbar puncture is clearly indicated (Figure 1). First, anyone who has syphilis and any neurologic sign or symptom needs a lumbar puncture. The second and third categories involve patients who are neurologically asymptomatic. The first of these groups consists of patients with evidence of tertiary syphilis -- such as cardiovascular syphilis or late skin manifestations of syphilis. These patients should have a lumbar puncture, but this category of patients is extremely rare in the antibiotic era.
The other category of patients without neurologic symptoms who should probably undergo a lumbar puncture consists of patients who get the appropriate treatment for syphilis and don't respond serologically to that treatment. If you rule out reinfection in these patients, they should probably undergo a lumbar puncture to make sure they don't have underlying asymptomatic neurosyphilis.
If you treated a patient for syphilis and their serological RPR titer has not declined, and if they say they had unprotected sex again and they noticed a lesion, you know they got reinfected and they don't need a lumbar puncture. But if a treated patient comes back with titers that didn't decline appropriately or even increased, and if that person says they've had no sex partner since starting syphilis therapy, they should undergo a lumbar puncture.
We know that patients with HIV and a CD4 count less than 350, or patients with HIV and syphilis and an RPR titer greater than or equal to 1:32, appear to be at increased risk of asymptomatic neurosyphilis. The problem is that in the antibiotic era we don't have any data to show that doing a lumbar puncture on these individuals ultimately improves their outcomes. It doesn't mean that lumbar puncture will not improve their outcome; it just means that we don't have any data. Because of that, and because there are risks associated with a lumbar puncture, the CDC recommendations aim to highlight the issue and not to make any formal recommendations beyond explaining when lumbar puncture should be considered.1
I'll tell you what I do: If I have an HIV-positive patient who is completely asymptomatic neurologically but has either a low CD4 count or a high RPR titer, I try to base my decision about lumbar puncture on how reliable that patient seems. If I think the patient is reliable and likely to come back for follow-up to make sure that they're neurologically well, or if they're likely to call me should any neurologic symptoms develop, I'm less inclined to do a lumbar puncture. But if I have a patient who's shown clearly that they're not reliable, they're not going to call, I try to get a lumbar puncture just to make sure I rule out asymptomatic neurosyphilis. It's not a perfect approach because sometimes it's hard to tell who's going to be reliable and who's not. But at the same time it's very difficult for us to schedule lumbar punctures on all our HIV patients. It's just not feasible.
Mascolini: Do syphilis treatment response rates in people with HIV differ from rates in the general population?
Ghanem: Several studies, including the only randomized trial addressing this issue,2 show that HIV-positive patients are slightly less likely to respond serologically to syphilis treatment than people without HIV. [See "Syphilis treatment response in people with HIV" in the review article, "Slowing resurgent syphilis in people with HIV."] That's why the CDC recommends close follow-up of HIV-positive patients treated for syphilis. To me, that is the single most important recommendation the CDC makes on syphilis because close follow-up lets you quickly identify individuals who don't respond appropriately to therapy. When you identify a poor response, you can retreat those patients or work them up more fully. Closer follow-up of treated patients would prevent many of the complications that could occur in poor treatment responders.
The problem is that it's hard to follow these patients up. For example, when we looked at rates of followup for HIV-positive patients in the Baltimore City Health Department, we found that about 65% of our patients treated for syphilis don't have a followup RPR titer in the next year.3 And we're confident in our results because we have access to RPR titers done throughout the state of Maryland. That's a huge number. I think close follow-up of treated patients is the most important thing we can do because it allows us to identify patients who are not responding adequately in a timely manner.
Mascolini: In your systematic review of syphilis treatment in people with HIV,4 you conclude that "guideline recommendations in this population are based on little objective data." How do you advise HIV clinicians to treat patients diagnosed with syphilis?
Ghanem: We found that the published data are limited in terms of suggesting the best syphilis treatment approach for our HIV-infected patients. We didn't intend to cause clinicians anxiety by stating this, because we have decades of experience in treating syphilis in people with HIV. What that experience tells us is that the vast majority of our patients coinfected with HIV and syphilis do very well on standard therapy.
Physicians can take heart in knowing that and in following the CDC recommendations, which state that the treatment of syphilis is virtually identical in HIV-infected and uninfected patients.1 The only difference is that HIV-infected patients should be followed up more aggressively after treatment. If we do that we could quickly identify the small subset of patients who don't respond well to therapy and treat them more aggressively.
Mascolini: Can you summarize your research on how antiretroviral therapy affects syphilis treatment response?
Four Common Clinical Mistakes in Managing Syphilis in HIV-Positive Patients
Ghanem: We looked at our HIV cohort at Hopkins to see whether the immunologic status of our HIV-infected patients had an impact on the course of syphilis.5 We found that patients whose immunological status was impaired -- in other words, those whose CD4 count was less than 350 -- had an increased risk of neurosyphilis if they had syphilis. We also found that they were less likely to respond as well serologically to syphilis therapy than patients whose CD4 count was above 350. And we found that use of highly active antiretroviral therapy tended to improve responses to syphilis therapy in these patients and to reduce the risk of neurological complications.
These findings were not really surprising because we already knew that successful treatment of syphilis depends on two things: You need a functioning immune system, and you need antibiotics. One without the other is not enough to control syphilis. In the pre-antibiotic era most syphilis patients had a good immune system, but we didn't have any effective drugs, and it was hard to manage syphilis. In the pre-HIV antibiotic era, we had patients with a good immune system and we had good drugs; as a result, the number of syphilis cases dropped and people did very well with syphilis treatment. Then came the HIV era. We still had good drugs for syphilis, but in people with HIV we didn't have an intact immune system. Clearly, what this tells us is that to manage syphilis we need good drugs and a good immune system. Clinician mistakes in managing syphilis
Mascolini: What are the most common mistakes HIV clinicians make in managing HIV-positive people with syphilis?
Ghanem: There are several things I would focus on. The first, and I think the most important, is not screening for syphilis nearly enough. Clinicians need to screen high-risk patients more frequently.
The second problem is not being aggressive in following up HIV-infected patients after they get treated for syphilis. You really want to bring them back in after treatment to make sure they're responding appropriately serologically.
Another thing clinicians tend to forget can be illustrated by this scenario: An HIV-infected patient comes in with early syphilis. They get the appropriate treatment. A week or 2 weeks later they call and say they're having headaches. Some clinicians don't think about the possibility of neurosyphilis because they think they've treated the syphilis and neurosyphilis can't occur after treatment. Clinicians tend to forget that neurosyphilis can occur after appropriate syphilis therapy, particularly in HIV-infected patients.
That's another reason why it's important to follow up with patients and to explain to patients what to watch for after they get treated for early syphilis: "If you develop a headache, if you notice visual changes, if you develop any neurologic function abnormality, give me a call immediately." Our study of HIV-positive people with syphilis showed that neurosyphilis can develop in those patients after they are appropriately treated for early syphilis.6
A final mistake involves the issue of early neurosyphilis. Clinicians tend to forget that early neurosyphilis does occur. They assume that neurosyphilis develops many years after the initial infection. But particularly among patients who have an immune system defect, such as people with HIV, early neurosyphilis can occur literally within days of infection. In other words you can have neurosyphilis concomitantly with primary syphilis, secondary syphilis, or early latent syphilis. In fact in our study early neurosyphilis was far more common than late neurosyphilis.6
Clinicians should remember that early neurosyphilis is something they have to look for. Whenever they're evaluating an HIV-infected patient for early syphilis, clinicians have to ask about neurological symptoms including photophobia, headache, neck stiffness, visual changes, and cranial nerve abnormalities. Doing so will help ensure that they don't miss a case of early neurosyphilis.
- Centers for Disease Control and Prevention. Sexually Transmitted Diseases Treatment Guidelines, 2010. MMWR Morb Mortal Wkly Rep. 2010;59:RR-12. Accessed February 23, 2012.
- Rolfs RT, Joesoef MR, Hendershot EF, et al. A randomized trial of enhanced therapy for early syphilis in patients with and without human immunodeficiency virus infection. The Syphilis and HIV Study Group. N Engl J Med. 1997;337:307-314. Accessed April 24, 2012.
- Ghanem KG, Erbelding EJ, Wiener ZS, Rompalo AM. Serological response to syphilis treatment in HIV-positive and HIV-negative patients attending sexually transmitted diseases clinics. Sex Transm Infect. 2007;83:97-101. Accessed April 24, 2012.
- Blank LJ, Rompalo AM, Erbelding EJ, Zenilman JM, Ghanem KG. Treatment of syphilis in HIV-infected subjects: a systematic review of the literature. Sex Transm Infect. 2011;87:9-16.
- Ghanem KG, Moore RD, Rompalo AM, Erbelding EJ, Zenilman JM, Gebo KA. Antiretroviral therapy is associated with reduced serologic failure rates for syphilis among HIV-infected patients. Clin Infect Dis. 2008;47:258-265. Accessed April 24, 2012.
- Ghanem KG, Moore RD, Rompalo AM, Erbelding EJ, Zenilman JM, Gebo KA. Neurosyphilis in a clinical cohort of HIV-1-infected patients. AIDS. 2008;22:1145-1151.