Search tips
Search criteria 


Logo of brjopthalBritish Journal of OphthalmologyVisit this articleSubmit a manuscriptReceive email alertsContact usBMJ
Br J Ophthalmol. 2007 September; 91(9): 1206–1208.
Published online 2007 May 2. doi:  10.1136/bjo.2007.117523
PMCID: PMC1954888

Diagnostic approaches for oculoglandular tularemia: advantages of PCR



The authors describe a diagnostic approach that proved to be particularly valuable in rare cases of ocular tularemia registered during the tularemia outbreak in 1997–2005 in Bulgaria. The authors describe the laboratory findings and diagnosis of four cases with an oculoglandular form of infection.


Several different specimens from each patient were analysed. Oculoglandular tularemia was diagnosed in four patients either by culture, immunofluorescent antibody analysis (IFA), serology or by a polymerase chain reaction (PCR) assay.

Results and Discussion

Three F tularensis strains were isolated and characterised. One of these was isolated from a conjuctival swab specimen obtained from a seronegative patient. The authors report for the first time a successful application of diagnostic PCR performed directly on conjuctival swab specimen. From all analysed specimens IFA was diagnostically effective only in the case of lymph node aspirates and was not sensitive enough for conjuctival swabs or blood samples. The authors also describe the histological picture of a conjunctival granuloma in course of infection. All patients were successfully treated with ciprofloxacin.


Some of the proposed laboratory diagnostic strategies (swab PCR) are not invasive and could represent a new approach for resolving rare and hard‐to‐diagnose cases of oculoglandular tularemia.

Tularemia is a zoonotic disease caused by the gram‐negative bacterium Francisella tularensis. It affects a variety of mammalian species, mostly humans, hares and rodents and survives in protozoa.1,2 The Francisellaceae family is in the γ subclass of Proteobacteria and range over closely‐related microorganisms incorporated within a single genus Francisella with two species F tularensis and F philomiragia.3F tularensis with its four subspecies (ssp tularensis, ssp holarctica, ssp mediasiatica and ssp novicida) is the causative agent of tularemia disease in North America, Europe, Asia and recently Australia.4F tularensis ssp tularensis is highly virulent (infectious dose less than 10 CFU) for various mammals including humans and is considered by the Centers for Disease Control and Prevention (CDC) a category A potential bio‐weapon agent. The genetic basis of its high virulence and pathogenicity is currently unknown. The second species F philomiragia is an opportunistic pathogen, rarely causing disease in mainly immunocompromised patients and often associated with water.

Depending on the route of entry, tularemia occurs in several clinical forms: glandular, ulceroglandular, oculoglandular, oropharyngeal, intestinal, pneumonic and typhoidal. Some authors accept only two forms—ulceroglandular (local) and typhoid (septic).5 The oculoglandular form of tularemia is one of the rarest ones comprising up to only 4% of all cases.6,7 Humans may acquire this disease form by direct contact with infected materials, often via contaminated hands or by aerosolised contaminated particles. The intracellular bacterium appears to enter and replicate in macrophages via a cytoplasm‐B‐insensitive pathway and without triggering a respiratory burst.8 The subsequent clinical presentation, known as Parinaud oculoglandular syndrome, is a rare medical condition characterised by granulomatous conjunctivitis, associated with homolateral cervical and/or pre‐ or retroauricular lymph adenopathy.9 It has been associated with several infectious diseases including tularemia which is rarely considered in the differential diagnosis.10

Herein we discuss our findings in four patients with Parinaud syndrome diagnosed during a tularemia outbreak.


All tularemia cases were diagnosed according to the case definitions of the CDC.11 Clinically relevant information was gathered by interview, referral to hospitals, and questionnaires and sent to general practitioners in the region for submission to the reference centres for epidemiological analysis. Informed consent was obtained from the subjects after explanation of the nature and possible consequences of the study, and the overall study was in accordance with the Declaration of Helsinki. Serum samples for serological assays for specific anti‐Francisella antibodies were collected from all patients 1–4 weeks after the onset of the disease (S1). Second and third samples were obtained after 14 days (S2) and three months later respectively (S3). All samples were tested with a serum tube agglutination and haemagglutination kit (BulBio‐NCIPD, Bulgaria). Serial dilutions of sera, starting from 1:20, and antigen prepared from F tularensis strain Srebarna 19 were mixed and after overnight incubation at 37°C the agglutination was evaluated by unaided eye. Serum specimens with [gt-or-equal, slanted]1:80 or at least fourfold titre rise in S2 or S3 samples were interpreted as positive results.12,13

Biopsy specimens from cervical lymph nodes (2 patients), conjunctival swab (1 patient), surgically extracted conjunctival granuloma (1 patient), and 3 ml blood samples (4 patients) were collected for culture and PCR detection.

Half of the volume from each specimen suspension was cultured on modified Thayer–Martin agar plates (containing Gc medium base, haemoglobin and IsoVitaLex at 37°C with 5% CO2 in air) and the other half was processed for PCR.4 All manipulations were carried out in a biosafety level 3 cabinet. DNA from biopsy specimens/aspirates (500 μl), blood samples (1.5 ml) or swabs was isolated following standard phenol‐chloroform protocol with slight modifications.14 DNA (100 ng) was used in PCR with tul4 and RD1 F tularensis specific primers according to the protocols by Johansson and Broekhuijsen.15,16

Direct immunofluorescence assay (IFA) with FITC‐conjugated anti‐ Francisella sera (BulBio‐NCIPD) was used to detect F tularensis antigens in all specimens except blood. Preparations (smears) from conjuctival lymph nodule (patient 3) and smears were finally counterstained with 0.25% Evans blue.

The histological techniques were as follows: briefly, 4% p‐formaldehyde was used for fixation and 4 μm paraffin embedded tissue sections were microscopically examined.


A typical clinical picture of oculoglandular tularemia was observed in all patients (fig 11).

figure bj117523.f1
Figure 1 Picture of a tularemia infected eye (patient 1). Informed consent was obtained for publication of this figure.

In addition to the Parinaud syndrome, two patients showed signs of cervical lymphadenopathy. IFA was positive only in the lymph node specimens but no visible fluorescence was detected in the swab smears. Except for one patient significant and/or rising agglutination titres were detected in all the other patients.

Surgically extracted conjuctival nodule with dimensions 5×2 mm was white‐yellow in colour. We observed an area surrounded by stratified non‐keratinised, squamous epithelium with regions of parakeratosis and submembrane manifested mononuclear infiltrations, pronounced vascularisation and single Langhans cells. A granuloma with central necrosis was detected in the centre of the nodule.

The systematic observations of all other diagnostic findings are summarised in table 11.

Table thumbnail
Table 1 Diagnostic findings observed in four Parinaud syndrome patients

Direct bacteriological isolation was achieved only from cervical lymph node aspirates from two patients. We did not succeed with attempts for primary direct isolation of the causative agent from the conjuctival swab and blood specimens. However, F tularensis was isolated from conjuctival swab suspension (500 μl) in one patient by intraperitoneal inoculation in a guinea pig and subsequent culturing of spleen material. All animals were treated in compliance with the Guide for the Care and Use of Laboratory Animals.17F tularensis was detected by PCR in both types of ocular specimens—conjuctival swab (patient 4) and surgically extracted conjunctival nodule (patient 3). The blood samples from these patients were also PCR positive (fig 22).). With the RD1 subspecies specific primers all isolates produced characteristic electrophoretic bands (900 bp) for ssp holarctica (fig 22).). In three cases, the consecutive serological samples exhibited marked titre dynamics, which further supported diagnosis. However, in patient 4 seroconversion did not take place and she remained seronegative for six months (when the last sample was tested) after successful therapy. This case had a clinical history of an extensive abdominal intervention shortly before the infection took place, and was resolved by positive PCR from conjuctival swab and blood but also with consecutive isolation of the causative agent (see table 11).

figure bj117523.f2
Figure 2 PCR on clinical specimens with tul4 and RD1 primers. (A) (1) F tularensis ssp holarctica strain LVS; (2) negative control, ddH2O; (3) lymph node, patient 1; (4) lymph node, patient 2; (5) blood, patient 3; (6) blood, patient 4; (7) conjuctival ...

All patients were successfully treated with ciprofloxacin 500 mg twice daily for 21 days. The rationale for this type of treatment was the specific drug pharmacodynamics characterised by good tear mucosal excretion.


A large number of studies discussing tularemia outbreaks in various regions have been published in the last 15 years.8 There are several publications describing outbreaks in the Balkan Peninsula with significant healthcare importance. The first ocular case was reported in Bulgaria in 2003.16

As the ocular forms are extremely rare and the overall index of suspicion is low, they often remain unresolved and under‐reported.18 The clinical picture of Parinaud syndrome is non‐indicative for a particular causative agent and more often is associated with Bartonella henselae, Mycobacterium tuberculosis or Herpes infections.

To our knowledge the direct detection and/or isolation of F tularensis from conjuctival swab material have not been reported in the literature. In the last decade PCR assays have proved to be of great clinical and diagnostic value, providing rapid and definitive results. Several such assays have been published for the detection of tularemia in ulcers, blood, ticks and so on but none for diagnosing ocular tularemia. Because of the requirements for special biosafety conditions when handling F tularensis and the very low isolation rate (~ 5%), cultures are not routinely performed in most of the laboratories. The fastidious nature of the microorganism prevents its detection in the common microbiological laboratories where specific media are not routinely applied. In these cases the Parinaud syndrome remains “idiopathic” with no aetiological diagnosis. With respect of tularemia the specific antibodies are not detectable in the early stages of the disease (up to the 14th day). In such cases the PCR assays performed either with ulcerous or conjuctival swabs are very useful alternatives for early confirmation of clinical diagnosis and proper treatment. Our diagnostic approach concerning swab analysis is not invasive or traumatic and appears to be highly effective and straightforward. In addition, it is supportive in cases where no specific immune response appears or is delayed. This was clearly demonstrated in one of our patients (patient 4), having no positive serology at the onset of disease or afterwards. The patient had a history of recently performed serious abdominal surgical intervention that had resulted in a significant level of immunosuppression.

These data suggest that we should strongly recommend a broader application of this methodology when oculoglandular tularemia is suspected.


CDC - Centers for Disease Control and Prevention


This work was in part supported by the Bulgarian Ministry of Education and Sciences, under project L‐MU‐1413.

Competing interests: None.

Informed consent was obtained for publication of figure 1.


1. Dennis D, Inglesby T, Henderson D. et al Tularemia as a biological weapon: medical and public health management. JAMA 2001. 2852763–2773.2773 [PubMed]
2. Kantardjiev T, Velinov T. Interaction between protozoa and microorganisms of the genus Franacisella. Problems of Infectious Diseases 1995. 2234–35.35
3. Sjöstedt A. Family XVII. Francisellaceae, genus I. Francisella. In: Brenner DJ, ed, Bergey's manual of systematic bacteriology. New York, NY: Springer 2003. 111–135.135
4. Sandstrőm G, Sjőstedt A, Forsman M. et al Characterization and classification of strains of Francisella tularensis isolated in the Central Asian focus of the Soviet Union and Japan. J Clin Microbiol 1992. 30172–175.175 [PMC free article] [PubMed]
5. Staples J E, Kubota K A, Chalcraft L G. et al Epidemiologic and Molecular Analysis of Human Tularemia, United States, 1964–2004. Emerg Infect Dis 2006. 71113–1118.1118 [PubMed]
6. Kantardjiev T, Ivanov I, Velinov T. et al Tularemia outbreak, Bulgaria, 1997–2005. Emerg Infect Dis 2006. 4678–680.680 [PubMed]
7. Fortier A. Life and death of an intracellular pathogen: Francisella tularensis and the macrophage. Immunol Ser 1994. 60349–361.361 [PubMed]
8. Steinemann T L, Sheikholeslami M R, Brown H H. et al Oculoglandular tularemia. Arch Ophthalmol 1999. 117132–133.133 [PubMed]
9. Starck T, Madsen B W. Positive polymerase chain reaction and histology with borderline serology in Parinaud's oculoglandular syndrome. Cornea 2002. 6625–627.627 [PubMed]
10. Centers for Disease Control and Prevention Division of Public Health Surveillance and Informatics Available from (accessed 20 February 2000)
11. Bevanger L, Maeland J, Naess A. Agglutinins and antibodies to Francisella tularensis outer membrane antigens in the early diagnosis of diseases during an outbreak of tularemia. J Clin Microbiol 1988. 26433–437.437 [PMC free article] [PubMed]
12. Olsufjev N, Tularemia Methods for laboratory diagnostics of tularemia. In: Olsufjev N, ed. Diagnostics of infections at high medical risk. Rostow University 1970. 170–179.179
13. Ausbel F M, Brent R, Kingston R E. et alCurrent protocols in molecular biology. John Wiley and Sons 1995. 2.4.1
14. Johansson A, Berglund L, Eriksson U. et al Comparative analysis of PCR versus culture for diagnosis of ulceroglandular tularemia. J Clin Microbiol 2000. 3822–26.26 [PMC free article] [PubMed]
15. Broekhuijsen M, Larsson P, Johanssonsen A. et al Genome‐wide DNA microarray analysis of Francisella tularensis strains demonstrates extensive genetic conservation within species but identifies regions that are unique to the highly virulent F. tulatensis subsp. tularensis. J Clin Microb 2003. 412924–2931.2931
16. Vassileva P, Petrow P, Kantardjiev T. et al Tularemia: first cases of oculoglandular form of the disease in Bulgaria. Comptes Rendus ‐ Academie Bulgare des Sciences 2003. 56117–120.120
17. Alberts B, Liebowitz H. Guide for the care and use of laboratory animals. Washington, DC: Institute of Laboratory Animal Resources Commission on Life Sciences, National Research Council, National Academy Press, 1996
18. Martin G J, Marty A M. Clinicopathologic aspects of bacterial agents. Clin Lab Med 2001. 221513–548.548 [PubMed]

Articles from The British Journal of Ophthalmology are provided here courtesy of BMJ Group