Journal of Neuro-Ophthalmology, March 2000, Volume 20, Issue 1

Annual Update of Systemic Disease

Journal of Neuro- Ophlhalmology 20( 1): 48- 58, 2000. © 2000 Lippincott Williams & Wilkins, Inc., Philadelphia Annual Update of Systemic Disease- 1999: Emerging and Re- emerging Infections ( Part II) Larry Frohman, MD, and Paul Lama, MD \* In Part 1 of this edition of the annual review of sys­temic diseases, which appeared in the 19: 4 ( December 1999) issue, we discussed Whipple's disease as an emerging infection and reviewed new findings in an es­tablished infectious disease, Lyme disease, that can ef­fect the visual system. In Part II, we discuss the emerging infections babesiosis, ehrlichiosis, and Hantavirus and will review new findings that can affect the visual system from syphilis, an established infectious disease. SYPHILIS Syphilis is caused by infection with the spirochete Treponema pallidum. It is a complex systemic illness and, as a result of its protean manifestations, has been long been dubbed the " Great Imitator." Thus, it has been said that to know syphilis is to know medicine. Unlike the spirochete, which causes Lyme disease, the mode of transmission is primarily by sexual contact; protective immunity does not occur, and reinfection is possible. Spirochetes are highly infectious. Horizontal transmis­sion, in addition to the common sexual route of trans­mission, may occur by nonsexual contact with lesions on an infected person, blood transfusion, or handling in­fected material. Vertical infection occurs via placental transmission in infected pregnant individuals. As few as 57 spirochetes may result in human infection, as deter­mined by intradermal infection in volunteers ( 81,82). An inoculum containing approximately 500 to 1,000 spiro­chetes is then present on average in naturally acquired infection based on a replication rate of 30 to 33 hours and an incubation period of 21 days before the clinical ap­pearance of a lesion. The natural history of syphilis in­volves three stages. The primary stage is characterized by the development of the well- known skin lesion, the chancre. The chancre is painless but may be tender to touch and is associated with regional lymphadenopathy. Chancres heal spontaneously 2 to 8 weeks later. The secondary stage is characterized by widespread dissemi- Manuscript received July 21, 1999; accepted November 1, 1999. From the Departments of Ophthalmology ( LF, PL) and Neurosci-ences ( LF), UMDNJ- New Jersey Medical School, Newark, New Jersey. Address correspondence and reprint requests to Larry Frohman, MD, 90 Bergen Street, Sixth Floor, Newark, NJ 07105. nation of the spirochete occurring 2 to 12 weeks after infection. This stage may manifest with diffuse mucocu­taneous, constitutional, and parenchymal involvement. The greatest numbers of treponemes are present at this stage, and spirochetes can be identified in a number of tissues including blood. The central nervous system ( CNS) and eye may be involved, and spirochetes may also be found in CSF and aqueous. High antigen loads are responsible for manifestations such as membranous nephropathy with nephrotic syndrome. This stage may be followed by a period of latency and finally a tertiary stage that occurs in approximately one of three untreated patients. Late syphilis is characterized by CNS, cardiac, and skeletal disorders. Common sites of gummatous parenchymal involvement are the liver, spleen, skin, and bones. The latent stage is characterized by an early latent ( within 4 years of contact) and late latent stage ( greater than 4 years) depending on whether relapses are possible. Seventy- five percent of relapses occur within the first year of infection. Ophthalmic and Neuroophthalmic Manifestations Ocular and neuroophthalmic manifestations are well known and include interstitial keratitis, scleritis, uveitis, vasculitis, neuroretinitis, optic neuritis, and chorioretini­tis. Involvement of the pupillary pathway manifests as an Argyll Robertson pupil. Involvement of any cranial nerve may occur leading to extraocular movement ab­normalities and diplopia. With human immunodeficiency virus ( HIV) co-infection, the number of reported cases of syphilis has increased to epidemic proportions. Associated with the rise in syphilis cases are increased reports of unusual manifestations, related to the immunocompromised state. In addition, treatment failure in patients with late mani­festations has become an issue of increasing concern in HIV- infected patients. Kuo et al. ( 83) described a case series of three HIV-positive patients who developed dense vitritis. All three patients had severe reduction in vision to hand motions, two had bilateral involvement, all had anterior uveitis characterized by moderate cell and flare with keratic pre­cipitates on the endothelium, and dense vitreous haze with obscuration of fundus details. Serologic testing dis- ANNUAL UPDATE OF SYSTEMIC DISEASE: PART II 49 closed marked elevations in rapid plasma reagin ( RPR) titers. All three had abnormal cerebrospinal fluid ( CSF) with positive white cells and positive CSF VDRL con­sistent with neurosyphilis. All patients demonstrated a positive response to intravenous penicillin therapy with improvement in visual acuity and reduction of the vitri-tis. Two of three patients had marked improvement in acuity within weeks of treatment. Although uveitis is the most common ocular manifestation of syphilis, primary vitritis without retinal vasculitis or retinitis is distinctly unusual. These cases highlight several important points: syphilis should always be included in the differential diagnosis of ocular inflammation, ocular treponemal in­fection by definition implies CNS syphilis, thus treat­ment should follow the protocol for neurosyphilis, and HIV status needs to be investigated in any patient who presents with ocular syphilis ( 83). Although syphilitic uveitis is nongranulomatous, Fon-seca et al. ( 84) reported a case in which the initial diag­nosis was thought to be leprosy. This patient had uveitis with widespread cutaneous eruptions. Histopathologic evaluation of the skin lesion revealed a dermal granulo­matous infiltrate with perineural invasion. He was ini­tially treated for leprosy, but subsequent testing revealed serology consistent with syphilis and concomitant HIV infection. Penicillin therapy was instituted with success­ful resolution of the uveitis and skin lesions. Whether this manifestation is unique to those with concomitant HIV infection remains to be determined. It seems likely, however, that those HIV- infected individuals capable of forming granulomas have higher CD4 counts. Although it is widely believed that interstitial keratitis ( IK) is most commonly luetic in etiology, a retrospective study reported by Schwartz et al. ( 85) strongly chal­lenged this belief. The records of 97 patients classified as having either active or inactive IK were reviewed. Spe­cifically, the patients' ocular examination, medical his­tory, and laboratory evaluation were analyzed. They found that Herpes simplex accounted for 71.4% and Va­ricella zoster accounted for 8.6% of cases of active uni­lateral IK. Idiopathic causes accounted for 14.3%. In patients with unilateral inactive IK, herpes accounted for 50.0% and 33.3% were idiopathic. Sixty percent of bi­lateral active IK cases was idiopathic; however, 48.5% of bilateral inactive cases was syphilitic in origin. Overall, syphilis accounted for only 18.6% of all cases. Thus, with the exception of bilateral inactive IK, most cases are herpetic in origin or idiopathic. Neurologic Manifestations A substantial portion of the recent literature on syphi­lis infection is related to late- stage neurosyphilis with emphasis on diagnosis and treatment in HIV- positive pa­tients. However, there is a number of interesting case reports describing unusual manifestations in nonimmu-nocompromised individuals. Seizure disorder presenting in middle adult life is usually related to a structural CNS lesion such as a tumor or an arteriovenous malformation. The frequency of seizures in neurosyphilis ranges from 14% to 60%. However, status epilepticus as the present­ing manifestation of late- stage syphilis is exceedingly rare. Primavera et al. ( 86) described a case of status epilepticus in a 44- year- old HIV- negative man with no history of seizure disorder. He presented with acute al­teration in mental status that was later attributed to de novo complex partial nonconvulsive status epilepticus. Results of lumbar puncture, neuroimaging, and the pa­tient's clinical course after antibiotic therapy were con­sistent with syphilitic meningovasculitis. Angus et al. ( 87) reported a case of temporal lobe symptoms without motoneuron signs associated with confusion in a 34- year- old man. Although such findings are consistent with herpes simplex encephalitis, espe­cially when associated with a significant red blood cell count in the CSF, this patient had positive CSF serology for syphilis. Ahmad and Lee ( 88) described a case of bilateral sensorineural hearing loss, vertigo, and progres­sive neurologic deficits in a 75- year- old patient. A sero­logic workup disclosed positive syphilis serology, but magnetic resonance imaging ( MRI) revealed multiple in­tracranial lesions. Although, type 2 neurofibromatosis was a consideration, with the positive syphilis serology, the lesions were believed to be syphilitic gummas. Thus, the patient was treated with penicillin for neurosyphilis. However, despite treatment, the patient eventually ex­pired as a result of progressive encephalopathy. Diagnosis In 1991, there were more than 128,000 new cases of syphilis reported to the Centers for Disease Control ( CDC). The acquired immunodeficiency syndrome ( AIDS) epidemic has significantly contributed to this re­surgence of syphilis. Factors that increase the risk of HIV infection also increase the risk of syphilis infection and include multiple sexual partners, illicit drug use, ex­change of sex for drugs, poor access to health care, and socioeconomic decline. Hence, co- infection with HIV and syphilis is very common. In particular, there is a number of reports suggesting a resurgence of neuro­syphilis in the HIV- infected population ( 89). Earlier neurosyphilis manifestations appear to be more prevalent than later stage manifestations such as tabes dorsalis or general paresis. This may be in part due to the reduced life span of the HIV patient, thus making the later manifestations less prevalent. Flood et al. ( 90) per­formed a retrospective analysis of all cases of neuro­syphilis presenting in San Francisco hospitals between 1985 and 1992. They identified 117 cases. Overall, 75% were HIV positive. A total of 65% had early symptom­atic ( 33%) or asymptomatic ( 32%) neurosyphilis. The earlier neurosyphilis syndromes were classified as acute syphilitic meningitis, meningovascular syphilis, and uve­itis. Eighty- nine percent of the cases of early neurosyphi­lis was HIV positive or had high- risk behavior for HIV infection. The most frequent manifestation of menin­govascular syphilis was hemiparesis, accompanied by dysarthria or a cranial nerve defect. The patients with uveitis presented with visual loss, and in two cases the uveitis was unilateral. Only 6% had late parenchymal neurosyphilis. The mean serum VDRL titer was reactive J Neuro- Ophlhalmol, Vol. 20, No. I, 2000 L. FROHMAN AND P. LAMA • i ,,*•' i''-'" /'"'' 3 » ; at a dilution of 1: 128. The highest median titers were in the patients with uveitis ( 1: 512). Overall serum titers were no different between HIV- positive and HIV-negative cases. A similar retrospective analysis was car­ried out in Sao Paulo, Brazil, by de Souza and Nitrini ( 91). Although there were fewer total numbers of pa­tients, they also found that the earlier manifestations pre­dominated in the HIV- positive cohort. In summary, com­pared to the cases of neurosyphilis in the preantibiotic era, the patients were younger and HIV positive, with earlier manifestations. The diagnosis of syphilis can be made serologically or by direct examination. Unlike most other pathogens, tre-ponemes cannot be cultivated in vitro. Direct examina­tion via dark- field examination or phase- contrast micros­copy of biopsy specimens or the serous transudate from a chancre, for example, can identify the organisms. Se-rodiagnosis of syphilis begins with serum identification of nontreponemal antibodies ( VDRL or RPR). A positive serum sample can flocculate a suspension of cardiolipin-cholesterol- lecithin antigen. Although the antigen is no longer obtained from beef heart or beef livers, neverthe­less these antibodies are nonspecific and false- positive reactions do occur as in patients with collagen vascular disease, HIV infection, hepatitis C, atypical pneumonia, and endocarditis. The sensitivity of syphilis serology in­creases with increasing stage of the illness except in the latent or tertiary stage, in which it may be absent or even untreated. Once a positive nontreponemal test is ob­tained, a specific treponemal antibody assay should be performed. There are currently three commonly per­formed specific treponemal tests: fluorescent treponemal antibody absorbed test ( FTA- abs), T. pallidum hemag­glutination assay ( TPHA, MHA- TP), and the treponemal immobilization test ( TPI). The sensitivity in early disease is highest for the FTA- abs at 85% and lowest for the TPI at 50%. The sensitivity increases to nearly 100% in those with secondary syphilis for the FTA- abs and MHA- TP and 97% for the TPI assay but decreases slightly in late-stage syphilis for all three tests. When evaluating a pa­tient with suspected syphilis, both nontreponemal and specific treponemal tests should be ordered simulta­neously as a negative nontreponemal assay certainly does not rule out syphilis infection especially in those with late- stage manifestations such as ocular, neurologic, or cardiac symptoms. Unlike Lyme disease serologies, syphilis serologies are more sensitive and specific; how­ever, false negatives have been reported, especially in those patients with dual syphilitic and HIV infection sus­pected of having late- stage disease. The diagnosis of neurosyphilis may be even more difficult as CSF VDRL is fairly insensitive. Thus, maintaining a high index of suspicion is critical, especially in those at high risk for syphilis infection. This was demonstrated in a study by Erbelding et al. ( 92) who assessed the serologic status of 1,117 intravenous drug users. In their study, they were able to identify 10% with syphilis and 10% with biologic false- positive ( BFP) VDRL. Sixty- one percent of these BFP reactors had dilutions that did not rise above 1: 8, and 31% had titers greater than 1: 8 dilutions, but the FTA- abs was always nonreactive. It was in the latter subgroup that three patients were found to have positive immunoblots to a specific 17- kDa T. pallidum membrane protein. Thus, these patients who ordinarily would be misclassified as having a BFP VDRL actually had syphi­lis infection. All three patients were HIV positive. A fourth HIV- positive patient also had a positive immuno-blot reaction but, in contrast, later developed a positive FTA- abs reaction. Such findings would certainly tempt clinicians to use this assay to identify true syphilis in­fection in those patients with positive nontreponemal as­says but negative FTA- abs tests, especially those who are HIV positive. However, close inspection of the statistics reveals that the apparent sensitivity of the FTA- abs test in the study by Erbelding et al. was 96.9%, which is not significantly different than the sensitivity of the FTA- abs assay reported in the pre- AIDS era ( 93). Hence, this study should not lead to indiscriminate testing as most HIV- positive patients with a positive VDRL and nega­tive FTA- abs are true BFP reactors as expected second­ary to polyclonal immunoglobulin activation. If the cli­nician is confronted with a patient at high risk for syphi­lis infection and the clinical manifestations are suggestive but initial testing reveals a negative FTA- abs, then further serologic testing using the 17- kDa immuno-blot assay may be indicated. Although the FTA- abs is highly specific for syphilis, false positives occur in 1% of cases. Those with chronic inflammatory diseases such as rheumatoid arthritis or biliary cirrhosis may have a false- positive FTA- abs as may patients with Lyme disease. The test is also positive in patients with nonvenereal treponemal infection ( yaws, pinta, and bejel). Despite the protean manifestations of syphilis, indiscriminate laboratory testing of patients at lower risk for infection such as those with other inflam­matory conditions or those who originate from areas en­demic for nonvenereal treponemal infection can lead to inappropriate assignment of the syphilis diagnosis. The diagnosis of CNS syphilis can be made several ways. A positive CSF VDRL always indicates neuro­syphilis. Patients with reactive blood serologies, CSF pleocytosis with elevated protein concentration, and compatible neurologic signs and symptoms, even in the absence of a reactive CSF VDRL, are also considered to have neurosyphilis. This is especially true in later forms of neurosyphilis such as tabes dorsalis, in which a reac­tive CSF VDRL is often lacking. What about those pa­tients with ocular or neurologic signs and symptoms, risk factors for syphilis, with or without concomitant HIV infection, who have nonreactive serologies in blood and CSF but with elevated CSF protein and pleocytosis? What is the role of polymerase chain reaction ( PCR) in the detection of T. palliduml In early- stage disease, bi­opsy of suspicious lesions followed by dark- field exami­nation or rabbit infectivity testing can easily identify T. pallidum. Such lesions of primary and secondary syphilis contain large numbers of spirochetes. The numbers of spirochetes, as mentioned previously, increase from pri­mary to secondary syphilis but begin to decline in later J Neuro- Ophthalmol. Vol 20, No. 1, 2000 ANNUAL UPDATE OF SYSTEMIC DISEASE: PART II 51 stages, thus making direct identification in late- stage dis­ease very difficult. In fact, the granulomatous lesions of syphilitic gumma are thought to be immune reactions to treponemal antigens in the absence of live spirochetes. Zoechling et al. ( 94) were able to identify T pallidum in five biopsy specimens using the highly sensitive tech­nique of PCR. All five specimens were biopsy negative using conventional direct identification techniques. One of the specimens was from a patient with gummatous syphilis. Thus, PCR may be useful in those situations in which serology is negative, as in neurosyphilis, HIV in­fection, and congenital syphilis. More experience is nec­essary before final recommendations on its usage in vari­ous clinical scenarios is forthcoming. Treatment The treatment of syphilis in its various stages has al­ways been under intense scientific scrutiny. This is es­pecially true since the start of the HIV era. The following recommendations represent a compilation of the avail­able clinical data via a MEDLINE search of the English-language literature. A single intramuscular dose of benzathine penicillin G is highly successful in eradicating primary and secondary lesions as well as permitting serologic recovery and thus has been the recommended treatment for stage I, II, and early latent syphilis. However, it has been shown that CNS invasion with spirochetes occurs in as many as 40% of patients during the early stages of syphilis infection, yet benzathine penicillin G does not provide measurable drug levels in CSF. Furthermore, since there have been anecdotal reports of neurosyphilis and other syphilis complications in patients who had been treated with ben­zathine penicillin G, investigators have been prompted to assess longitudinally the efficacy of benzathine penicillin G. To date, there is only one published study that ad­dressed this issue. Rolfs et al. ( 95) compared standard therapy with benzathine penicillin G alone, plus it com­bined with a 10- day course of amoxicillin with probeni-cid in 541 patients with early syphilis. Nineteen percent of these were HIV positive. The results indicated that, with regard to resolution of chancres and rashes, there was no significant difference between standard versus enhanced therapy independent of HIV status. Although serologic failure was more common in the HIV- positive group, it did not differ according to treatment group. There was only one treatment failure and that occurred in an HIV- positive patient. Thus, enhanced therapy did not improve clinical or serologic outcome, and HIV- positive patients respond less well serologically, regardless of treatment. It is interesting to note that 25% of these pa­tients had T. pallidum detected in the CSF before treat­ment. This indicates that early invasion of the CSF is a poor predictor of treatment failure. It may be that spiro­chetes in early disease move in and out of the CSF and high sustained serum treponemicidal levels may prevent persistent CSF infection from occurring. Longer follow-up, however, is necessary to determine whether there is a greater rate of development of neurosyphilis in the standard treatment arm of the study compared to en­hanced therapy. At present, the recommendation for treatment of early- stage syphilis remains 2.4 million units of intramuscular benzathine penicillin G even in HIV- positive patients. Treatment for late latent syphilis is longer. The current recommendation is weekly intra­muscular injections of benzathine penicillin G for 3 weeks. The rationale behind this is that spirochetes di­vide slowly in late disease and thus longer courses are necessary. Augenbraum and Rolfs ( 96) compiled the re­sults of articles pertaining to syphilis treatment since the last CDC revision of recommendations on the treatment of sexually transmitted diseases and answered many im­portant treatment related questions. With respect to treat­ment of late latent syphilis, including those patients in­fected with HIV, the data reveal that the current recom­mendation appears to be adequate in both groups of patients. However, they do recommend CSF analysis in all HIV- infected patients as the rate of asymptomatic neurosyphilis is high. Those with asymptomatic neuro­syphilis should be treated accordingly, which is 2 weeks of 2 to 4 million units of intravenous aqueous penicillin every 4 hours. As of 1998, the treatment guidelines for established neurosyphilis is 18 to 24 million units of aqueous penicillin daily for 10 to 14 days. The greatest controversy regarding this regimen, as expected, is in the HIV- infected subgroup. Malone et al. ( 97) reported a 70% serologic failure in HIV- positive patients. The study, however, was small and no comparisons were made with HIV- negative patients. Overall, HIV- positive patients have a slower rate of decline of serologic mark­ers in serum and CSF after therapy. Persistent serum markers, however, may indicate the presence of a BFP VDRL. Therefore, patients with neurosyphilis and HIV infection need to be followed very carefully if serologic markers are not responding to appropriate therapy or symptoms persist; then repeated courses of therapy are indicated. They recommend treatment at the higher range of doses regardless of whether the patients are asymp­tomatic or symptomatic. Malessa et al. ( 98) pointed out in their study of syphilitic HIV- positive patients with latent disease that, regardless of CSF VDRL status, those with CSF abnormalities should be treated for asymptom­atic neurosyphilis. Since penicillin is the antibiotic of choice in all stages of syphilis, what then is the role of other antibiotics such as tetracyclines, macrolides, or ceftriaxone? Are there any reasonable alternatives to penicillin? At present, there is no established role for any of these antibiotics, especially in the late stages of syphilis. Although there are anecdotal reports of ceftriaxone, there are unfortu­nately no hard data supporting or refuting its use in any stage of syphilis infection. Azithromycin 500 mg once daily has been tried in a clinical trial of patients with early syphilis and found to be successful in curing 11 of 13 ( 85%) patients ( 99). De Maria et al. ( 100) evaluated the efficacy of minocycline in a small series of three HIV- negative men allergic to penicillin. They received minocycline 100 mg twice daily for 2 weeks every month for 9 months. Clinical signs and symptoms im- J Neuro- Ophlhalmol, Vol. 20, No. 1, 2000 52 L. FROHMAN AND P. LAMA proved after 1 month. CSF VDRL became nonreactive in two patients by 6 months, with the disappearance of CSF pleocytosis at 6, 12, and 24 months. The third patient did not complete the intended course of therapy. Although there are reports citing alternative antibiotics for the treatment of syphilis, large prospective trials with these agents are lacking. In those with late- stage disease and penicillin allergy, penicillin desensitization is thus rec­ommended. Minocycline may be used in those with peni­cillin allergy and early- stage syphilis. Once the diagnosis of syphilis is established and peni­cillin therapy is instituted, serologic parameters must be followed to assess the efficacy of the regimen. Nontre-ponemal antibody titers decrease with response to treat­ment, whereas specific treponemal antibodies often per­sist indefinitely. Those patients with rising or persistent titers and/ or persistent symptoms are easily identified as treatment failures. However, those with declining titers are expected to do so at a specific rate, otherwise they also would be classified as treatment failures. Precisely what constitutes an appropriate rate of decline has been redefined. Previously a fourfold decline in titer was ex­pected within 3 months of treatment in early- stage syphi­lis, but this has since been changed to a 6- to 12- month period, hence obviating the need for retreatment or re-evaluation until approximately 1 year in those with slower rates of decline. HIV- infected patients have a slower rate of decline compared to the HIV- negative population and often nontreponemal antibodies persist indefinitely. The more advanced the stage of syphilis, the longer the elapsed time interval until disappearance of antibodies. In those with persistent low- grade titers, dis­tinguishing between persistent infection and a BFP VDRL would be very difficult. Hence, close clinical ob­servation and monitoring of other parameters, such as CSF white blood cell count and protein in those with neurosyphilis, for example, would aid in establishing whether the treatment is effective. BABESIOSIS Babesiosis is another spirochetal infection, first de­scribed in Europe in 1957. It is transmitted by tick bites. Clinically it resembles malaria, with headache, fever, chills, nausea, vomiting, myalgia, altered mental status, disseminated intravascular coagulation, anemia, hy­potension, respiratory distress, and renal insufficiency as clinical features of both entities ( 101). Babesiosis may also present as a fever of unknown origin associated with anemia ( 102). The symptoms of babesiosis are typically more severe in the asplenic or immunocompromised ( 103). Kim et al. ( 104) recently reported two new symptoms of babesiosis: bradycardia and lymphopenia. In Europe, 83% of infections are in asplenic individu­als, the most common organism was Babesia divergens, which is a cattle parasite. A newer antigenic form, MOl, has also been associated with human infection. The dis­ease seems to be more common in the United States, with many cases infected by ticks carrying the rodent parasite B. microti. Several such cases appear to have been the result of blood transfusion. These B. microti cases tend to be less severe than B. divergens infections. Cases of babesiosis can exist despite patients being seronegative to B. microti, with newer forms emerging ( 103,105,106). Unfortunately, epidemic babesiosis infection may also occur as a result of transfusion products derived from an asymptomatically infected individual ( 107). White et al. ( 108) reviewed the experience in the state of New York with 139 cases of babesiosis hospitalized between 1982 and 1993. They reported that 91% of these cases lived in Suffolk County ( Long Island), 62% were male, and the mean age at initial hospitalization was 62.5 years. Nine patients ( 6.5%) died, 35 ( 25.2%) required admission to the intensive care unit, 52% of patients had a history of chronic babesiosis, 12% had a history of Lyme disease, 12% had undergone a splenectomy, and 2% had undergone a blood transfusion. Babesiosis, like malaria, has typically been detected by examining peripheral blood smears. Krause et al. ( 109) studied an immunoglobulin M ( IgM) indirect im-munofluorescent antibody assay for acute babesiosis. They found that the test was 91% sensitive and 99% specific and was a reliable test for the routine clinical diagnosis of acute babesiosis. Krause et al. also studied the duration of infection as determined by DNA analysis in residents of two endemic areas, Block Island and southeastern Connecticut. They found that silent sub­clinical infection was not rare. In the seroconverted pa­tients who did not mount symptoms, the mean duration of detection of babesial DNA was 82 days. In patients who had clinical symptoms, babesial DNA persisted for 16 days in those who received therapy ( clindamycin and quinine). They concluded that untreated silent babesial infection may persist for months or even years ( 110). Babesiosis is not infrequently coincident with Lyme infection. This may be more common in endemic areas where both organisms are carried by the same tick vec­tor, Ixodes scapularis. Illness severity and duration typi­cally are greater in those patients with co- infection ( 111). In a study of 1,156 subjects in Connecticut and Rhode Island, Krause et al. ( 112) found 97 ( 8.4%) had serologic evidence of Lyme disease. Among these 97, 14 ( 14%) also had reactivity against babesial antigen. Patients who had both illnesses had a statistically higher incidence of conjunctivitis, headache, as well as fatigue, sweats, chills, anorexia, emotional lability, nausea, and spleno­megaly than did those with only Lyme disease. They also tended to have a longer duration of their symptoms. Thus, in areas where both diseases are endemic, one needs to suspicious of dual infection. Neurologic and ophthalmic involvement in babesiosis is not common. Nonspecific signs such as photophobia, conjunctival in­jection, and headache may occur. More severe and focal neurologic defects are usually secondary to severe sys­temic illness in cases with renal and hepatic disease. Seizures may occur, and ischemic and hemorrhagic strokes can be seen. Cornblath ( 113) reported a case of anterior ischemic optic neuropathy, presumably related J Neuro- Ophthalmol, Vol. 20, No. 1. 2000 ANNUAL UPDATE OF SYSTEMIC DISEASE: PART II 53 to severe hemolytic anemia, that was part of the clinical picture. Retinal cotton wool spots and soft exudates were reported ( 114,115). The classic therapy for babesiosis is quinine and clin­damycin. In their study of DNA evidence of babesial infection, Krause et al. ( 110) concluded that although treatment with clindamycin and quinine may reduce the duration of infection, the therapy was not benign, and relapse did occur. Thus, subclinical infection ( i. e., sero­conversion) need not be treated. Shih and Wang ( 116) reported the use of azithromycin in a patient who failed standard therapy. They concluded that azithromycin plus quinine may be a reasonable alternative regimen for ba­besiosis. A therapy that has also been used is massive exchange transfusion followed by clindamycin ( 103). Evenson et al. ( 117) believe that more severely ill or immune- suppressed patients with babesiosis should be considered for red blood cell exchange and/ or plasma­pheresis. EHRLICHIOSIS An Israeli study by Harrus et al. ( 118) demonstrated that ehrlichiosis can cause blindness. In their patient, the acute presentation of monocytic ehrlichiosis was acute blindness with bilateral hyphema, retinal hemorrhage, and detachment. The organism, however, was Ehrlichia canis and their patient a Labrador retriever. Thus, it is difficult to extrapolate the significance of their findings of ocular vasculitis to potential human ocular disease. Human ehrlichiosis is a tick- borne disease caused by rickettsial- like members of genus Ehrlichia. It was first described in 1994, and by 1998, more than 300 cases had been described ( 119). Two types of ehrlichiosis are seen in humans in the United States: human monocytic ehrli­chiosis ( HME) and human granulocytic ehrlichiosis ( HGE). Human monocytic ehrlichiosis is caused by E. chaffeensis, which targets mononuclear phagocytes. Hu­man granulocytic ehrlichiosis is caused by a member of the genus whose DNA strongly resembles E. equi. As it is yet unnamed, it is referred to in the literature as the " HGE agent." Human granulocytic ehrlichiosis is the third recognized human disease transmitted by /. scapu-laris ticks ( Lyme and babesiosis being the other two) and infects granulocytes. Most cases of ehrlichiosis are either asymptomatic or mild and occur approximately 4 to 10 days after a tick bite. It probably requires as long as 36 hours of exposure to the tick for infection to occur ( 120). The spectrum of illness is similar to babesiosis and malaria, with both forms featuring fever, headache, mal­aise, nausea, vomiting, myalgia, and anorexia. Physical findings are not common, with fever and skin rash being the most common ( 121- 123). Less common manifesta­tions include hepatosplenomegaly, pancytopenia ( 124), and cough ( 125), seen in one third of cases, and diarrhea ( 126). Human granulocytic ehrlichiosis may be fatal, as in the case of Jahangir et al. ( 127), in which the patient succumbed to pericarditis. In a series from the state of New York that identified 18 adults with HGE, the most common presenting symp­toms were fever ( 94%) and myalgia or arthralgia ( 78%). The most typical laboratory abnormalities were leukope­nia, thrombocytopenia, and abnormal liver enzymes. Thirteen patients ( 71 %) remembered a tick bite antedat­ing their illness. The illness was relatively mild, five patients ( 28%) were briefly hospitalized, and none died. All responded to doxycycline ( 128). The disease is not restricted to adults, and its clinical features are somewhat different in children. Schutze and Jacobs ( 129) reported a series of cases of HME in 12 children. The clinical spectrum of the disease was similar to what is commonly seen in adults with the exception that eight of 12 ( 67%) had developed hyponatremia. Al­though the series was small, it suggested that the disease may have a more fulminant course in children, as four of 12 ( 33%) of patients presented in shock and three ( 25%) required blood pressure support and mechanical ventila­tion. The long- term morbidity that was seen included a decrease in cognitive and neurologic performance. Exposure without clinical disease is not rare. Bakken et al. ( 130) surveilled 475 Wisconsin residents without clinical history suggestive of ehrlichiosis using an indi­rect immunofluorescence assay. They found that 15% of these residents tested positive and concluded that sub­clinical infection was not rare in Wisconsin. In a study performed in Bulgaria, Christova and Dumler ( 131) found that immunofluorescent evidence of exposure to the HGE agent was seen in 2.9% of blood donors and 4% of healthy patients who had been exposed to ticks. The United States National Center for Infectious Dis­eases recently reviewed its experience as a referral center for serologic testing for HGE. They detected cases of HGE from 21 states; 60% of the cases occurred in Con­necticut, New York, and Wisconsin, with cases occurring from April through December. The number of cases peaked in June and July ( 132). Bakken et al. ( 122) found that greater disease severity was seen with increased age, anemia, increased percent­age of neutrophils and decreased percentage of lympho­cytes in peripheral blood, and the presence of morulae ( the characteristic clusters of bacteria forming mem­brane- bound intracellular inclusions) in neutrophil cyto­plasm. In this series, two of 41 ( 5%) died from their illness. Neurologic Illness Nonspecific neurologic symptoms were reported in HME and HGE; these include headache, altered mental status, hyperactive reflexes, and seizures. Focal neuro­logic signs are rare in ehrlichiosis. The cases of neuro­logic involvement are better studied in HME; these pa­tients typically have lymphocytic CSF, pleocytosis, and elevated CSF protein ( 123,133). Horowitz et al. ( 134) reported the first case of involve­ment of the peripheral nervous system in HGE. This man developed fevers 1 week after a tick bite. Despite treat­ment with several antibiotic regimens ( including cefop-erazone, erythromycin, and cefaclor), he developed bi­lateral brachial plexopathies several weeks into his course of what was later shown to be HGE. This symp- J Neuro- Ophthalmol, Vol. 20, No. 1, 2000 54 L. FROHMAN AND P. LAMA torn persisted despite 3 weeks of therapy with doxycy-cline. The authors postulated that this symptom is due to an immune- mediated response and suggested that HGE must be included in the differential diagnosis of periph­eral nervous system disease in patients who have expo­sure within an endemic area. Grant et al. ( 135) reported a case of HME with mul­tiple organ system involvement, including significant neurologic disease. This 54- year- old man presented with flu- like symptoms of fever, chills, malaise, and a stiff neck. As he did not improve, it was noted that his hepatic transaminase levels were rising ( another sign of ehrli­chiosis). He developed nonlocalizing neurologic signs including headache, meningismus, and altered mental status. Focal findings included unilateral arm weakness and a Bell's palsy. The MRI scan only showed a nonen-hancing lesion in right frontal lobe subcortical white matter, which was bright on T2 and was thought to be an infarct. CSF showed elevated protein and IgG, 19 white cells, with 78% plasma cells and plasmacytoid lympho­cytes; all cultures were negative. A bone marrow biopsy showed a typical lesion for HME, the fibrin- ringed granuloma. Brain and meningeal biopsy showed atypical lymphoid cells in the leptomeninges with a predilection for blood vessel wall involvement. The patient gradually responded to intravenous doxycycline. Demyelinating polyneuropathy may also occur. Bakken et al. ( 136) reported a 48- year- old woman who presented with the typical flu- like syndrome of myalgia, fever, chills, and severe headache. She had experienced recent tick bites. Her physical examination was unre­markable, no rashes were noted, and she was begun on doxycycline. She rapidly developed bilateral facial numbness, dysarthria, and symmetrical extremity weak­ness. Other notable findings were pupils that were 5 mm and only sluggishly reactive to light, complete ophthal­moplegia, left- sided ptosis, right facial nerve paresis, symmetrical generalized weakness, and perioral anesthe­sia. The reflexes were hypoactive and the plantar re­sponses equivocal. MRI scan showed only diffuse en­hancement of the meninges with gadolinium. The CSF showed a normal opening pressure with a protein level of 1,160 mg/ dL, a glucose level of 38 mg/ dL ( serum = 90 mg/ dL), and 59 white cells ( 42% granulocytes). Intrave­nous ceftriaxone was added, and all CSF cultures were negative. All testing for Lyme disease was negative. Al­though PCR of the CSF was negative for the HGE agent, a PCR on blood was positive for this bacteria. Six months into her course, neurologic symptoms still per­sisted. Thus, it is clear that in endemic areas, ehrlichiosis must be considered as a possible etiologic agent in the clinical picture of Guillain- Barre and Miller Fisher syn­dromes. Ratnasamy et al. ( 137) reviewed the neurologic in­volvement in 57 confirmed cases of human ehrlichiosis. Fifteen of these cases had sampling of CSF, eight of 15 ( 53%) were abnormal, with lymphocytic pleocytosis and elevated protein being the most common findings. They identified 21 other cases of human ehrlichiosis in the literature in which CSF had been analyzed, 13 ( 62%) of these were abnormal, and four of 21 ( 19%) patients with CNS manifestations of ehrlichiosis and abnormal CSF findings died. A case of specific neuroophthalmic fi'ndings was re­ported by Carter and Miller ( 138). This 44- year- old pa­tient with HME developed a trochlear nerve paresis with lymphocytic meningitis. This woman sustained a tick bite and 3 to 4 weeks later developed headaches, nausea, fever, and myalgia. Despite therapy with cefuroxime for a presumed viral syndrome, she did not improve, and inpatient evaluation revealed elevated antibodies to E. chaffeensis at a titer of 1: 160 on an immunofluorescent assay ( IFA), although IgM was negative. The Lyme titer was negative; treatment was intravenous cefotaxime and ceftriaxone, and then oral doxycycline. Approximately 1 week after admission, diplopia developed, and neuro­logic assessment was performed. The MRI scan was nor­mal; the spinal fluid showed 240 white blood cells/ mm3, with 83%. lymphocytes. The protein level was slightly elevated at 58 mg/ dL. Neuroophthalmic assessment showed the diplopia to be due to a superior oblique palsy. A repeat spinal tap included assessment for syphilis, Lyme, and ehrlichiosis, all of which were negative, the serum titer for Ehrlichia reverted to normal, and the dip­lopia resolved over several months. Cranial neuropathy remains a distinctly rare manifes­tation of human ehrlichiosis. Diagnosis As with many infectious diseases, diagnostic tech­niques are, at the time of this writing, in rapid evolution. Bakken et al. ( 122) evaluated diagnostic modalities in an evaluation of 41 cases of HGE from Wisconsin and Min­nesota. These methods included examination of Wright-stained peripheral blood smears for the presence of neu­trophilic morulae, PCR of blood samples for Ehrlichia DNA, and evaluation of serologic response to the E. equi antigen by indirect immunofluorescent antibody assay. Leukopenia, anemia, and thrombocytopenia were com­mon nonspecific laboratory findings. Whereas only 16 of 37 ( 43%) of patients tested by PCR were positive, 80% of the patients so tested demonstrated morulae in the cytoplasm of peripheral blood neutrophils. Acute or con­valescent blood samples detected antibodies against E. equi in 38 of 40 ( 95%) of patients via IFA. They con­cluded that microscopic examination of the acute- phase blood smear for neutrophilic morulae was the most rapid and practical screening method for diagnosing HGE. Grant et al. ( 135) pointed out that these morulae may also be demonstrated within the CSF lymphocytes, fa­cilitating the diagnosis of neuro- HME. Everett et al. ( 123) reported that 20 of 23 ( 87%) of ehrlichiosis patients in their series had a positive PCR for E. chaffeensis sequences. In a series from New York State, PCR was positive in nine of 12 patients ( 75%), and the agent that causes HGE was cultured from two pa­tients. The antibody to E. equi was detected in 94% of cases ( 128). Ravyn et al. ( 139) reported the development J Neuro- Ophthalmol, Vol. 20. No. 1, 2000 ANNUAL UPDATE OF SYSTEMIC DISEASE: PART II 55 of an IFA, enzyme immunoassay ( EIA), and Western immunoblot using HL- 60 cell ( a line of human promy­elocyte leukemia cells) culture- derived human isolates. EIA was found to be superior to IFA. They recommend a two- stage strategy for identifying cases of HGE, with screening by EIA and confirming the specificity by Western blot. This strategy also pointed out the fre­quency of concomitant ehrlichiosis infection in patients with Lyme disease, as in using both EIA and Western blot, 20% of their patients with early Lyme disease had evidence of exposure to the agent causing HGE. Finally, Goodman et al. ( 140) reported on three cases of HGE who had fever, granulocytopenia, and thrombo­cytopenia with probable tick exposure. Although only one of three ( 33%) had neutrophilic inclusions ( moru-lae), PCR was positive for Ehrlichia in all three. In an attempt to culture the organisms, the patients' blood was inoculated into cultures of HL- 60 cells. In all three cases, infection of the cell line was documented by visualiza­tion of intracellular organisms and cell lysis. Subse­quently, it was confirmed by immunofluorescent micros­copy, PCR analysis, and DNA sequencing that the cul­tured organisms were Ehrlichia. This is the first report of culturing of the causative organism of HGE. Therapy Doxycycline is a reliable, well- tolerated therapy for ehrlichiosis ( 123). Aguero- Rosenfeld et al. ( 128) re­ported that all the patients in their series from New York responded to this therapy. Horowitz and Wormser ( 141) recently stated that this is the drug of choice for HGE and HME, with clinical response typically seen in 1 to 2 days. Weaver et al. ( 142), however, reported a patient who, despite being treated with doxycycline within 1 day of symptoms, still had a severe course characterized by gen­eralized edema, pulmonary infiltrates, acute respiratory distress syndrome, and noncardiogenic pulmonary edema. Patel and Byrd ( 143) reported a similar patient with ehrlichiosis who, despite early therapy with doxy­cycline, developed severe acute respiratory distress syn­drome that required institution of corticosteroids to con­trol. Weinstein ( 121) believes that rifampin or chloram­phenicol can be tried in cases in which tetracyclines are absolutely contraindicated. He also believes that due to the 5% to 10% mortality rate, treatment should never be delayed pending serologic or PCR confirmation of the diagnosis. Buitrago et al. ( 144) suggested that the use of rifampin be considered in infected patients who are preg­nant. As with Lyme disease and babesiosis, ehrlichiosis and Lyme disease may be coincident, as they may share a potential vector, the tick I. scapularis. Unlike in babe­siosis, in which this co- infection rarely appears to have serious implications, undetected co- infection with ehrli­chiosis may have dire clinical sequelae unless proper therapy is instituted. Various studies of coendemic areas in the United States showed that 2.2% to 26% of adult /. scapularis ticks were co- infected with both B. burgdor­feri and an agent resembling E. equi. Thus, one tick bite can yield both infections ( 125). Ahkee and Ramirez ( 145) reported a case of concurrent Lyme meningitis and ehrlichiosis in a patient with occupational exposure to ticks. The serologies in this case supported dual infec­tion. They concluded that empiric therapy for both Lyme disease and ehrlichiosis should be considered in any pa­tient suspected of having a tick- borne illness and dem­onstrating consistent signs and symptoms of both ill­nesses. This co- infection may not be a rare event. Magnarelli et al. ( 146) reported the results of serologic surveillance for HGE in Connecticut, an area that all the tick- borne zoonoses are endemic. In these patients, they simulta­neously assayed for E. equi, E. chaffeensis, B. microti, and B. burgdorferi. They found that 26% of patients had antibodies to two or more tick- borne agents. Wormser et al. ( 147) reported on their experience with evidence of co- infection in Westchester, New York, an endemic area for both ehrlichiosis and Lyme disease. In their first 10 cases of HGE, 100% showed antibody se-ropositivity to E. equi, and 70% had a positive PCR on whole blood. Although none of the 10 had any of the classic manifestations of Lyme disease ( cranial neurop­athy, erythema migrans, arthralgia), nine of 10 ( 90%) tested positive for Lyme disease using enzyme- linked immunosorbent assay ( ELISA) or immunoblot. Leuko­penia and thrombocytopenia may be markers for patients with Lyme disease who also have ehrlichial infection; unlike in Lyme disease, 70% of the patients in this series were leukopenic and 70% were thrombocytopenic ( in an unpublished series of these authors' patients with Lyme disease, only 2% had thrombocytopenia and 3% leuko­penia). However, the authors pointed out that it was as yet undetermined whether this is a true Borrelia co-infection, as opposed to a nonspecific false- positive as­say for Lyme in a patient with ehrlichial infection. They do believe that the data argue for use of a tetracycline agent as therapy in these cases, as it covers both organ­isms. To truly demonstrate that co- infection, as opposed to infection with one agent and a false- positive serology to the other agent, can occur would require the culturing of both agents from a single patient. This was recently achieved in a 47- year- old man who presented with non­specific symptoms ( myalgia, arthralgia, fever, headache, weakness, dizziness, difficulty concentrating, stiff neck, and cough). He had had a tick removed a month before he became ill without a rash having been noted at that time, although on presentation a rash consistent with er­ythema migrans was seen on his flank, as was right con­junctival injection. Notable laboratory tests that provided clues to dual infection were a low platelet count, a low white cell count, and elevated liver function assays. The patient rapidly responded to doxycycline 100 mg twice daily for 14 days. Evidence for HGE in this patient in­cluded demonstration of morulae in white cells, elevated serum titers via immunofluorescence against E. equi, a positive PCR for the HGE agent on whole blood, and J Neuro- Ophthalmol, Vol. 20, No. 1, 2000 56 L. FROHMAN AND P. LAMA demonstration of morulae in cultured cells infected with the patient's blood. Furthermore, the organism within the culture line was demonstrated by PCR and immunoflu­orescence to be the HGE agent. Evidence of the presence of Lyme infection included demonstration of motile spi­rochetes in a cultured skin biopsy specimen, shown by PCR to be B. burgdorferi, as well as a positive PCR test of the skin biopsy itself. ELISA and immunoblot of se­rum were also positive ( 125). Thus, in the patient with Lyme disease who has atypical findings such as throm­bocytopenia and leukocytopenia or liver function abnor­malities, consideration should be given to incorporating doxycycline into the therapeutic regimen while awaiting the results of definitive testing aimed at determining whether there is truly a dual infection present. HANTAVIRUS The epidemic outbreak of Hantavirus pulmonary syn­drome in the four corners region of the southwestern United States beginning in 1993 has led to an explo­sion of interest in the syndrome, with 197 of the 200 MEDLINE references on hantaviral disease at the time of this writing having been published since 1993. The illness featured a prodrome of fever and myalgia, fol­lowed by thrombocytopenia, the presence of immature white blood cells on the peripheral smear, and cata­strophic respiratory failure associated with the sudden onset of noncardiogenic pulmonary edema and hypoten­sive shock. The deer mouse, Peromyscus maniculatus, was quickly found to be the rodent reservoir for this virus ( 148). Le Guenno ( 149), however, pointed out that this is not a new pathogen, having been known in China for hun­dreds of years. It is also much more common in France than in the United States. Some strains have been known to cause a syndrome called hemorrhagic fever with renal syndrome in Eurasia, whereas other forms, particularly those seen in the United States, cause an acute respira­tory distress syndrome, now known as Hantavirus pul­monary syndrome, which has a high mortality rate ( 149,150). Chapman and Khabbaz ( 148) reported that this Hantavirus syndrome has been long known among the indigenous healers in the southwestern United States; records demonstrate at least three prior epidemics this century in the region. Blasi ( 151) reported that the mor­tality rate in Hantavirus pulmonary syndrome in 100 confirmed cases is 52%. Hantavirus is harbored in rodents; human- to- human transmission does not occur. Diagnosis is based on se­rologic assays. It must be stressed that exposure to aero­solized mouse droppings must be avoided, such as might be experienced in cleaning areas where domestic animals may be kept, especially in poorly ventilated areas. In Belgium, an epidemic began the year before the southwestern United States cases, and Colson et al. ( 152) reported on 62 serologically proven cases of the Neph-ropathia epidemica form of hantaviral infection. Fever was seen in all cases, headache in 71%, and abdominal or lumbar pains in 80%. Nonspecific respiratory symptoms were only seen in approximately 25% of cases. The most common laboratory abnormalities were thrombocytope­nia ( 69%), leukocytosis ( 77%), abnormal lactate dehy­drogenase level ( 69%); an elevation of C- reactive protein ( CRP) was not unusual. Abnormal renal function was seen in 84%. Liver enzymes were abnormal in 46%, stressing that one must include Hantavirus infection in the differential diagnosis of viral hepatitis. In this series, two neurologic signs were seen, one case of encephalitis and one case of Guillain- Barre were observed. They re­ported that acute myopia was seen in 24% and was thought to be a specific sign. Ahlm et al. ( 153) reported the neurologic and ophthal­mic findings in 26 prospectively followed patients with hemorrhagic fever with renal syndrome. Headache was seen in 96%, insomnia in 83%, and vertigo in 79%. Ocu­lar symptoms were reported by 82% of patients; 4 1% had photophobia and 50% reported blurred vision. One pa­tient had a seizure with transient loss of vision. MRI scans revealed white matter changes in approximately half of the cases. Finally, Lee et al. ( 154) reported a case in Korea with an isolated abducens nerve palsy as a complication of hemorrhagic fever with renal syndrome. Acknowledgment: Supported in part by an unrestricted grant form Research to Prevent Blindness, Inc., New York City, New York. REFERENCES 81. Bennett JC, Plum F. Cecil textbook of medicine. Philadelphia: W. B. Saunders, 1996. 82. Mandell GL, Bennett JE, Dolin R. Principles and practices of infectious diseases. 4th ed. New York: Churchill Livingstone, 1995. 83. Kuo IC, Kapusta MA, Rao NA. Vitritis as the primary manifes­tation of ocular syphilis in patients with HIV infection. Am J Ophthalmol 1998; 125: 306- 11. 84. Fonseca E, Garcia- Sllva J, del Pozo J, et al. Syphilis in an HIV infected patient misdiagnosed as leprosy. J Cutan Pathol 1999; 26: 51^ 1. 85. Schwartz GS, Harrison AR, Holland EJ. Etiology of immune stromal ( interstitial) keratitis. Comea 1998; 17: 278- 81. 86. Primavera A, Solaro C, Cocito L. De novo status epilepticus as the presenting sign of neurosyphilis. Epilepsia 1998; 39: 1367- 9. 87. Angus F, Maysuria H, Bryan CS. Neurosyphilis mimicking her­pes simplex encephalitis. J S C Med Assoc 1998; 94: 315- 7. 88. Ahmad I, Lee WC. Otoneurosyphilis masquerading as neurofi­bromatosis type II. ORL J Otorhinolaryngol Relat Spec 1999; 61: 37- 40. 89. Johns DK, Tierney M, Felsenstein D. Alteration in the natural history of neurosyphilis by concurrent infection with the human immunodeficiency virus. N Engl J Med 1987; 316: 1569- 72. 90. Flood JM, Weinstock HS, Guroy ME, et al. Neurosyphilis during the AIDS epidemic, San Francisco, 1985- 92. J Infect Dis 1998; 177: 931^ 10. 91. de Souza MC, Nitrini R. Effects of human immunodeficiency virus infection on the manifestations of neurosyphilis. Neurology 1997; 49: 893- 4. 92. Erbelding EJ, Vlahov D, Nelson KE. Syphilis serology in human immunodeficiency virus infection: evidence for false- negative fluorescent treponemal testing. J Infect Dis 1997; 176: 1397^ 100. 93. Sordillo EM, Hoehl B, Belch J. False- negative fluorescent trepo­nemal tests and confirmation of syphilis infection. J Infect Dis 1998; 1783: 294- 5. J Neuro- Ophthalmol, Vol. 20, No. 1. 2000 ANNUAL UPDATE OF SYSTEMIC DISEASE: PART II 57 94. Zoechling N, Schluepen EM, Soyer HP, et al. Molecular detection of Treponema pallidum in secondary and tertiary syphilis. Br J Dermatol 1997; 36: 683- 6. 95. 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. N Engl J Med 1997; 337: 307- 14. 96. Augenbraum MH, Rolfs R. Treatment of syphilis: nonpregnant adults. Clin Infect Dis 1999; 28: S21- 8. 97. Malone J, Wallace M, Hendrick B, et al. Syphilis and neurosyphi­lis in a human immunodeficiency virus type- 1 seropositive popu­lation; evidence for frequent serologic relapse after therapy. Am J Med 1995; 99: 55- 63. 98. Malessa R, Agelink MW, Hengge U, et al. Oligosymptomatic neurosyphilis with false- negative CSF- VDRL in HIV- infected individuals. Eur J Med Res 1996; 1: 299- 302. 99. Verdon M, Handsfield H, Johnson R, et al. Pilot study of azithro­mycin for the treatment of primary and secondary syphilis. Clin Infect Dis 1994; 19: 486- 8. 100. De Maria A, Solaro C, Abbruzzese M, Primavera A. Minocycline for symptomatic neurosyphilis in patients allergic to penicillin. N Engl J Med 1997; 337: 1322- 3. 101. Clark IA, Jacobson LS. Do babesiosis and malaria share a com­mon disease process? Ann Trop Med Parasitol 1998; 92: 483- 8. 102. Falagas ME, Klempner MS. Babesiosis in patients with AIDS: a chronic infection presenting as fever of unknown origin. Clin Infect Dis 1996; 22: 809- 12. 103. Gorenflot A, Moubri K, Precigout E, Carey B, Schetters TP. Human babesiosis. Ann Trop Med Parasitol 1998; 92: 489- 501. 104. Kim N, Rosenbaum GS, Cunha BA. Relative bradycardia and lymphopenia in patients with babesiosis. Clin Infect Dis 1998; 26: 1218- 9. 105. Herwaldt B, Persing DH, Precigout EA, et al. A fatal case of babesiosis in Missouri: identification of another piroplasm that infects humans. Ann Intern Med 1996; 124: 643- 50. 106. Uguen C, Girard L, Brasseur P, Leblay R. Human babesiosis in 1997. Rev Med Intern 1997; 18: 945- 51. 107. Dobroszycki J, Herwaldt BL, Boctor F, et al. A cluster of trans­fusion- associated babesiosis cases traced to a single asymptom­atic donor. JAMA 1999; 281: 927- 30. 108. White DJ, Talarico J, Chang HG, Birkhead GS, Heimberger T, Morse DL. Human babesiosis in New York State: review of 139 hospitalized cases and analysis of prognostic factors. Arch Intern Med 1998; 158: 2149- 54. 109. Krause PJ, Ryan R, Telford S 3rd, Persing D, Spielman A. Effi­cacy of immunoglobulin M serodiagnostic test for rapid diagnosis of acute babesiosis. J Clin Microbiol 1996; 34: 2014- 6. 110. Krause PJ, Spielman A, Telford SR 3rd, et al. Persistent parasit­emia after acute babesiosis. N Engl J Med 1998; 339: 160- 5. 111. Sweeney CJ, Ghassemi M, Agger WA, Persing DH. Coinfection with Babesia microti and Borrelia burgdorferi in a western Wis­consin resident. Mayo Clin Proc 1998; 73: 338^ » . 112. Krause PJ, Telford SR 3rd, Spielman A, et al. Concurrent Lyme disease and babesiosis. Evidence for increased severity and du­ration of illness. JAMA 1996; 275: 1657- 60. 113. Cornblath WT. Protozoa and protozoal diseases. In: Miller NT and Newman NJ, eds. Clinical neuro- ophthalmology, 5th ed. Bal­timore: Williams & Wilkins, 1998: 4613- 36. 114. Ortiz JM, Eagle RC Jr. Ocular findings in human babesiosis ( Nantucket fever). Am J Ophthalmol 1982; 93: 307- 11. 115. Zweifach PH, Shovlin J. Retinal nerve fiber layer infarct in a patient with babesiosis. Am J Ophthalmol 1991; 112: 597- 8. 116. Shih CM, Wang CC. Ability of azithromycin in combination with quinine for the elimination of babesial infection in humans. Am J Trop Med Hyg 1998; 59: 509- 12. 117. Evenson DA, Perry E, Kloster B, Hurley R, Stroncek DF. Thera­peutic apheresis for babesiosis. J Clin Apheresis 1998; 13: 32- 6. 118. Harrus S, Ofri R, Aizenberg I, Waner T. Acute blindness asso­ciated with monoclonal gammopathy induced by Ehrlichia canis infection. Vet Parasitol 1998; 78: 155- 60. 119. Bakken JS. The discovery of human granulocytotropic ehrlichio­sis. / Lab Clin Med 1998; 132: 175- 80. 120. Katavolos P, Armstrong PM, Dawson JE, Telford SR 3rd. Dura­tion of tick attachment required for transmission of granulocytic ehrlichiosis. J Infect Dis 1998; 177: 1422- 5. 121. Weinstein RS. Human ehrlichiosis. Am Earn Phys 1996; 54: 1971- 6. 122. Bakken JS, Krueth J, Wilson- Nordskog C, Tilden RL, Asanovich K, Dumler JS. Clinical and laboratory characteristics of human granulocytic ehrlichiosis. JAMA 1996; 275: 199'- 205. 123. Everett ED, Evans KA, Henry RB, McDonald G. Human ehrli­chiosis in adults after tick exposure. Diagnosis using polymerase chain reaction. Ann Intern Med 1994; 120: 730- 5. 124. Friedman AD, Daniel GK, Qureshi WA. Systemic ehrlichiosis presenting as progressive hepatosplenomegaly. S Med J 1997; 90: 656- 60. 125. Nadelman RB, Horowitz HW, Hsieh TC, et al. Simultaneous human granulocytic ehrlichiosis and Lyme borreliosis. N Engl J Med 1997; 337: 27- 30. 126. Devereaux CE. Human monocytic ehrlichiosis presenting as fe­brile diarrhea. J Clin Gastroenterol 1997; 25: 544- 5. 127. Jahangir A, Kolbert C, Edwards W, Mitchell P, Dumler JS, Pers­ing DH. Fatal pancarditis associated with human granulocytic ehrlichiosis in a 44- year- old man. Clin Infect Dis 1998; 27: 1424- 7. 128. Aguero- Rosenfeld ME, Horowitz HW, Wormser GP, et al. Hu­man granulocytic ehrlichiosis: a case series from a medical center in New York State. Ann Intern Med 1996; 125: 904- 8. 129. Schutze GE, Jacobs RF. Human monocytic ehrlichiosis in chil­dren. Pediatrics 1997; 100: E10. 130. Bakken JS, Goellner P, Van Etten M, et al. Seroprevalence of human granulocytic ehrlichiosis among permanent residents of northwestern Wisconsin. Clin Infect Dis 1998; 27: 1491- 6. 131. Christova IS, Dumler JS. Human granulocytic ehrlichiosis in Bul­garia. Am J Trop Med Hyg 1999; 60: 58- 61. 132. Comer JA, Nicholson WL, Olson JG, Childs JE. Serologic testing for human granulocytic ehrlichiosis at a national referral center. J Clin Microbiol 1999; 37: 558- 64. 133. Mathisen GE, Weiss PJ, Kennedy CA. Pneumonia, aseptic men­ingitis, and leukopenia in a 28- year- old man. Clin Infect Dis 1993; 16: 809- 15. 134. Horowitz HW, Marks SJ, Weintraub M, Dumler JS. Brachial plexopathy associated with human granulocytic ehrlichiosis. Neu­rology 1996; 46: 1026- 9. 135. Grant AC, Hunter S, Partin WC. A case of acute monocytic ehrlichiosis with prominent neurologic signs. Neurology 1997; 48: 1619- 23. 136. Bakken JS, Erlemeyer SA, Kanoff RJ, Silvestrini TC 2nd, Good­win DD, Dumler JS. Demyelinating polyneuropathy associated with human granulocytic ehrlichiosis. Clin Infect Dis 1998; 27: 1323- 4. 137. Ratnasamy N, Everett ED, Roland WE, McDonald G, Caldwell CW. Central nervous system manifestations of human ehrlichio­sis. Clin Infect Dis 1996; 23: 314- 9. 138. Carter N, Miller NR. Fourth nerve palsy caused by Ehrlichia chaffeensis. J Neuroophthalmol 1997; 17: 47- 50. 139. Ravyn MD, Goodman JL, Kodner CB, et al. Immunodiagnosis of human granulocytic ehrlichiosis by using culture- derived human isolates. J Clin Microbiol 1998; 36: 1480- 8. 140. Goodman JL, Nelson C, Vitale B, et al. Direct cultivation of the causative agent of human granulocytic ehrlichiosis. N Engl J Med 1996; 334: 209- 15. 141. Horowitz HW, Wormser GP. Doxycycline revisited: an old medi­cine for emerging diseases. Arch Intern Med 1998; 158: 192- 3. 142. Weaver RA, Virella G, Weaver A. Ehrlichiosis with severe pul­monary manifestations despite early treatment. S Med J 1999; 92: 336- 9. 143. Patel RG, Byrd MA. Near fatal acute respiratory distress syn­drome in a patient with human ehrlichiosis. S Med J 1999; 92: 333- 5. 144. Buitrago MI, Ijdo JW, Rinaudo P, et al. Human granulocytic ehrlichiosis during pregnancy treated successfully with rifampin. Clin Infect Dis 1998; 27: 213- 5. 145. Ahkee S, Ramirez J. A case of concurrent Lyme meningitis with ehrlichiosis. Scand J Infect Dis 1996; 28: 527- 8. 146. Magnarelli LA, Ijdo JW, Anderson JF, Padula SJ, Flavell RA, J Neuro- Ophthalmol, Vol. 20, No. 1, 2000 58 L. FROHMAN AND P. LAMA Fikrig E. Human exposure to a granulocytic Ehrlichia and other 151. Blasi F. New pathogens for respiratory infections. Curr Opin tick- borne agents in Connecticut. / Clin Microbiol 1998; 36: Pulm Med 1996; 2: 174- 80. 2823- 7. 152. Colson P, Damoiseaux P, Brisbois J, et al. Epidemic of Hantavi- 147. Wormser GP, Horowitz HW, Dumler JS, Schwartz I, Aguero- rus disease in Entre- Sambre- et- Meuse: year 1992- 1993. Clinical Rosenfeld M. False- positive Lyme disease serology in human and biological aspects. Acta Clin Belg 1995; 50: 197- 206. granulocytic ehrlichiosis. Lancet 1996; 347: 981- 2. 153. Ahlm C, Linden C, Linderholm M, et al. Central nervous system 148. Chapman LE, Khabbaz RF. Etiology and epidemiology of the and ophthalmic involvement in nephropathy epidemica ( Euro- Four Corners Hantavirus outbreak. Infect Agents Dis 1994; 3: 234- pean type of haemorrhagic fever with renal syndrome). J Infect 44. 1998; 36: 149- 55. 149. Le Guenno B. Hantavirus and pulmonary pathology. Rev Pneu- 154. Lee EY, Choi SO, Choi GB, Kang DH, Yoon KI. Isolated mol Clin 1998; 54: 393- 8. abducens nerve palsy as a complication of haemorrhagic 150. Schmaljohn C, Hjelle B. Hantaviruses: a global disease problem. fever with renal syndrome. Nephrol Dial Transplant 1998,13: Emerg Infect Dis 1997; 3: 95- 104. 2113- 4. < i 1\ • >. • i J Neuw- Ophthalmol, Vol. 20, No. I, 2000