1314

Identification of Leptospira Species in the Pathogenesis of Uveitis and

Determination of Clinical Ocular Characteristics in South India
Kathryn M. Chu, R. Rathinam, P. Namperumalsamy,
and Deborah Dean

Departments of Medicine and the Francis I. Proctor Foundation,

University of California at San Francisco, San Francisco, California,
and Aravind Eye Hospital, Madurai, India

Uveitis is considered a rare complication of leptospirosis. This report describes an epidemic of

uveitis among patients with leptospirosis and provides data, using polymerase chain reaction (PCR)
amplification of Leptospira DNA, that the pathogenesis is associated with anterior chamber spirochetes. Forty-six uveitis patients, 49 uveitis controls, and 54 cataract controls were enrolled at
Aravind Eye Hospital (Madurai, India). Leptospiral DNA was detected by PCR of aqueous humor;
serum antibody titers were determined by ELISA and microagglutination (MAT). Thirty-seven
uveitis patients (80%) demonstrated leptospiral DNA compared with 5 controls (8%; P .001).
Thirty-three uveitis patients (72%) had positive serology compared with 10 uveitis controls (20%)
and 13 cataract controls (24%; P .001). This report describes the largest cluster of patients with
leptospiral uveitis and identifies six clinical characteristics that provide a diagnostic profile for
leptospiral uveitis. This profile will be important for determining treatment regimens in countries
where PCR and MAT are not available.

Received 17 July 1997; revised 16 December 1997.

Presented in part: American Society for Microbiology Annual Meeting, New
Orleans, May 1996 (abstract D-46), and International Leptospirosis Society
Meeting, Nantes, France, September 1996.
Informed consent was obtained from all study participants according to the
guidelines of Aravind Eye Hospital, Madurai, India. Human experimentation
guidelines of the US Department of Health and Human Services and those of
Aravind Eye Hospital were followed in the conduct of this research.
Financial support: NIH (EY-00310 to D.D.). K.M.C. was supported in part
by the Deans Research Scholarship, University of California, San Francisco,
School of Medicine.
Reprints or correspondence: Dr. Deborah Dean, University of California,
San Francisco, School of Medicine, Box 0412, S-307, 513 Parnassus Ave.,
San Francisco, CA 94143-0412 (debd@itsa.ucsf.edu).
The Journal of Infectious Diseases 1998;177:131421
q 1998 by The University of Chicago. All rights reserved.
00221899/98/77050023$02.00

/ 9d45$$my18

03-03-98 14:07:02

jinfas

presented with pulmonary complications, including hemoptysis; 10% developed hepatic compromise. In a recent outbreak
in Nicaragua in 1995, 2000 patients developed an acute febrile illness and 40 died of pulmonary hemorrhage [5]. Leptospirosis was not initially suspected, because pulmonary complications had been rare [4, 6] and there was no evidence
of significant renal or hepatic involvement. It is possible that
different presentations of leptospirosis may impact the incidence and progression of other complications of this infection
(e.g., ocular disease).
Subconjunctival hemorrhage is the most common ocular
complication of leptospirosis, occurring in up to 92% of patients [7, 8]. Uveitis, which was first described in 1886 by Weil
[9], occurs in 2% 10% of patients and has been described as
self-limiting [2]. Uveitis can develop early or late in disease
and has been reported up to 1 year after the initial illness [2,
7, 10, 11]. The full spectrum of human ocular disease related
to leptospirosis is unknown, due to a low suspicion of ocular
involvement and lack of specific diagnostic tests.
In 1994, an increase in the number of individuals with uveitis
was noted at Aravind Eye Hospital, Madurai, India, after an
epidemic of leptospirosis in South India [12]. The epidemic
followed a severe flooding of Tamil Nadu District in the autumn of 1993. Some of these patients with uveitis presented
with a distinct constellation of signs, including panuveitis with
anterior chamber cell and flare, hypopyon, and vasculitis [12].
Further, all patients had elevated antibody titers to leptospiral
antigens by the microagglutination test (MAT). However, these
titers may have reflected systemic and not ocular disease.
We conducted a prospective case-control study to evaluate
two things: whether the presence of Leptospira DNA detected
by polymerase chain reaction (PCR) in the anterior chamber
is associated with the pathogenesis of uveitis, and whether

UC: J Infect

Downloaded from jid.oxfordjournals.org by guest on July 13, 2011

Leptospirosis is one of the most common zoonoses in the

world. The causative agent is the gram-negative spirochete
Leptospira; 200 species have been described [1]. The clinical
manifestations are varied. Leptospirosis usually presents as a
mild illness with fever, myalgia, and headaches but can progress to Weils syndrome, a severe, debilitating infection with
liver dysfunction, acute renal failure, and microvascular hemorrhage. The mortality rate from untreated Weils syndrome is
5% 30% [2]. In countries where leptospira are endemic, the
presentation of this disease is often confused with malaria,
tuberculosis, viral hepatitis, typhoid, and other parasitic and
bacterial infections. Indeed, only 30% of leptospirosis patients
are correctly diagnosed [3].
The spectrum of human disease caused by Leptospira species
is wide. In the Andaman Islands prior to 1931, the majority
of leptospirosis patients presented with Weils syndrome [4].
However, during outbreaks in 1993 and 1995, most patients

JID 1998;177 (May)

Leptospiral Uveitis

patients with uveitis and the clinical characteristics described

by Rathinam et al. [12] are caused by Leptospira species. We
determined whether MAT titers, PCR, or both are reliable predictors of ocular infection. MAT and PCR results were also
analyzed to identify characteristic (for leptospiral uveitis) ocular findings, which could be used in a clinical diagnostic algorithm in countries where these tests are not readily available.
Methods

cal examinations were performed to rule out pharyngitis, paresthesias, joint erythema or effusions, and hepatosplenomegaly. Ocular
histories included pain, redness, photophobia, floaters, and visual
deficit. Slit lamp examination and indirect ophthalmoscopy were
performed. Anterior chamber and fundus photographs were obtained to document each finding.
Individuals who earned Rs 2000/month were considered to
be of low socioeconomic status (SES).
Collection and handling of specimens. About 100200 mL of
aqueous humor was collected from one affected eye of each uveitis
patient and each uveitis control by anterior chamber paracentesis.
Aqueous humor was obtained from cataract controls during surgery. Blood (10 mL) was obtained for serologic testing. All specimens were stored at 0707C until further processing.
Serologic and hematologic testing. MAT was performed at
the World Health Organization Reference Laboratory, Brisbane,
Australia. Briefly, a library of 20 known reference serovars was
used, representing known pathogenic serovars. Each serum sample
was tested at a 1:50 dilution against pools of serovars. Reactive
sera were then subjected to serial 2-fold dilutions and reacted
against each serovar to determine the end-point titer. An ELISA
was also performed to determine the presence of IgM antibodies
as previously described [13].
All sera were tested for syphilis by VDRL test. Widal tests were
performed on all uveitis patients and controls to detect agglutinating antibodies to O and H antigens of Salmonella typhi. Thick
smears on whole blood were prepared to rule out malaria. All
VDRL, Widal, and malaria tests were performed and recorded by
the Microbiology Laboratory of Aravind in a blinded fashion.
Amplification of leptospiral DNA. Aqueous humor (50 mL)
was added to 25 mL of a solution of 50 mM Tris-Cl (ph 8.3), 1
mg/mL proteinase K, 0.45% Nonidet P-40, and 0.45% Tween 20.
All samples were handled in a blinded fashion. After incubation
at 557C for 2 h, the sample was incubated at 1007C for 10 min
and centrifuged at 14,000 g for 5 min; the supernatant was transferred to a DNase-free tube. DNA was amplified as previously
described [14], except that primers specific for Leptospira were
used. In brief, 5 mL of the supernatant was used in a 100-mL
reaction volume. The Leptospira primer pair was GF1 5*-CTG-

Table 1. Causes of uveitis among uveitis control patients.

Location of abnormality
Etiology

No. (%),
n 49

Anterior

Idiopathic*
Tuberculous
Fuchs
Viral
Sarcoidosis
HLA-B27
VKH
Posttrauma
Behcets
Leprosy

27
10
5
2
2
1
1
1
1
1

25
7
5
2
2
1
0
1
1
1

(55)
(20)
(10)
(4)
(4)
(2)
(2)
(2)
(2)
(2)

(51)
(14)
(10)
(4)
(4)
(2)
(2)
(2)
(2)

Posterior

Panuveitis

Bilateral

0
3 (6)
0
0
0
0
0
0
0
0

2 (4)
3 (6)
0
0
0
0
1 (2)
0
0
0

1 (2)
3 (6)
0
0
0
0
1 (2)
0
0
0

* All known infectious and testable diseases were ruled out.

/ 9d45$$my18

03-03-98 14:07:02

jinfas

UC: J Infect

Downloaded from jid.oxfordjournals.org by guest on July 13, 2011

Study population and case definitions. This study was conducted at Aravind Eye Hospital in 1995. Aravind is a major referral
hospital for the district of Tamil Nadu, which has a population of
100 million. About 1600 patients are seen each day, resulting
in an annual census of 600,000.
Patients 1865 years old from Tamil Nadu who consented were
consecutively enrolled. Uveitis cases (henceforth referred to as
uveitis patients) were defined as patients with four ocular findings
suggestive of leptospiral uveitis and included panuveitis, anterior
chamber cells, flare, and vasculitis. These four findings were used
to identify a uveitis patient because they had previously been found
among individuals thought by Rathinam et al. [12] to have leptospiral uveitis. Two age-matched control groups were established. A
5-year age range was used for matching: 1014 years, 1519,
2024, 2529, 3034, etc.
Uveitis controls were defined as individuals with active uveitis
with ocular presentations or known diagnoses distinct from the
uveitis patients or both (table 1), although these controls could
have one, two, or three of the four ocular findings used to define
a uveitis patient. Cataract controls were defined as individuals
without uveitis who presented for cataract surgery.
In addition, histories of leptospirosis (table 2) occurring during
the flood of 1993 or within 18 months of the event were obtained
for each uveitis patient and each uveitis and cataract control, and an
MAT titer 1:100 was considered indicative of infection. Medical
histories also included headache, myalgia, rash, chest pain, dry
cough, hemoptysis, upper quadrant abdominal pain, nausea, vomiting, diarrhea, constipation, and febrile illness in the family. Physi-

1315

1316

Chu et al.

JID 1998;177 (May)

Table 2. Comparison of systemic symptoms for uveitis patients and uveitis and cataract controls.
Uveitis patients
(n 46)

Characteristics
Fever
Myalgias
Headache
Rash
Hemoptysis
Interval to presentation
Average
Range

36
29
25
5
4

Uveitis controls
(n 49)

(78)
(63)
(54)
(11)
(9)

36
26
32
5
1

6.8 mo
7 d to 18 mo

Cataract controls
(n 54)

(73)
(53)
(79)
(10)
(2)

4.8 mo
4 d to 18 mo

32
25
12
3
0

(59)
(46)
(22)
(6)

P*
NS
NS
P .003
NS
NS

5.2 mo
5 d to 18 mo

NOTE. d, days; mo, months. Data are no. (%) unless stated otherwise.
* Comparison of uveitis patients with uveitis controls and independently with cataract controls. NS not significant.

Comparison of uveitis patients with cataract controls; there was no statistical significance between uveitis patients
and uveitis controls.

Interval between systemic illness and onset of ocular symptoms.

Results
Demographic characteristics. We enrolled 149 individuals
in the study: 46 uveitis patients, 49 uveitis controls, and 54

/ 9d45$$my18

03-03-98 14:07:02

jinfas

cataract controls (table 3). The average age of uveitis patients

was 33 years; most were from rural areas with low SES, and
many had occupations that involved animal contact.
Although an attempt was made to match, by sex, uveitis
patients with uveitis controls through consecutive enrollment,
we had a disproportionate number of male uveitis patients by
the end of the study. It has been well documented that there is
a higher prevalence of leptospirosis among male than female
persons in endemic regions [11], which may explain this finding.
Systemic findings. The most common symptoms described
for leptospirosis are fever, myalgias, headache, and rash. Although nonspecific, these findings were compared among uveitis patients and among uveitis and cataract controls (table 2).
The distribution of symptoms was similar for both uveitis
groups, but fewer individuals in the cataract group were symptomatic.
Of the uveitis patients, 22 (48%) had prior evidence for
leptospirosis compared with 10 (20%) for uveitis controls and
6 (10%) for cataract controls.
Serology and hematology results. All thick smears for malaria, VDRL, and Widal tests were negative for uveitis patients
and both control groups. Twenty-four uveitis patients (67%)
were seroreactive to leptospiral serovars by MAT (table 4); 14
(58%) of these 24 were reactive to 1 serovar. The uveitis
patients were nonreactive to serovars celledoni, zanoni, kremastos, cynopteri, ballum, bataviae, djasiman, and shermani.
Eight uveitis controls (16%) were seroreactive by MAT; 5
of them were reactive to 1 serovar. Uveitis controls were
seroreactive to 10 serovars, 9 of which were also seroreactive
in the uveitis patients. Seven cataract controls (13%) were
seroreactive to one serovar except for 1 individual who was
reactive to 2. Four of the 7 serovars were also serovars to
which uveitis patients were reactive.
Twenty-two uveitis patients (48%) were IgM-positive by
ELISA compared with 8 uveitis controls (16%; P .003) and
10 cataract controls (19%; P .004; table 5). Of the 24 uveitis

UC: J Infect

Downloaded from jid.oxfordjournals.org by guest on July 13, 2011

AATCGTGTATAAAAGT (bp 119) and GB2 5*-GGAAAACAAATGGTCGGAAG (bp 247266) [15]. The PCR thermocycling profile and PCR product verification were performed as
previously described [14].
Nested PCR was performed if a DNA band of the appropriate
size was not visualized. Thus, nested PCR was performed on all
samples that were negative by the first PCR. The first PCR product
(3 mL) was used to amplify a 174-bp region internal to GF1 and
GB2 using oligonucleotides LepF1 5*-CGAAATAAAATAACGCATGATACCAA (bp 5479) and LepB1 5*-AAAGTAAACGGCGCGAAC (bp 194211). The PCR reaction volume and thermocycling profile were as stated above.
PCR results were verified by duplication of the amplification
reactions as described above, using the original sample with both
primer pairs. There were no discrepant results.
Data analysis. The estimated sample size for uveitis patients
to detect a 10% difference (90%) between uveitis patients and
controls was calculated using a 0.05 and a power of 0.80. We
calculated a sample size of 45, which was sufficient to provide a
95% confidence interval that would be 10%. The target number
for each control group was 45.
Stratified analysis was performed using the Mantel-Haenszel
test, and multivariate analyses were performed using unconditional
logistic regression. For discrete data, the x2 and Fishers exact
tests were used.
The PCR and MAT/ELISA results for both uveitis patients and
uveitis controls were compared. In addition, the results for cataract controls were also compared with those from uveitis patients
and uveitis controls. Clinical ocular characteristics were compared between all PCR-negative and all PCR-positive persons
and between all PCR- and MAT/ELISA-positive and all PCRand MAT/ELISA-negative persons, regardless of which group
(uveitis patient, uveitis control, or cataract control) the individual
belonged to.

JID 1998;177 (May)

Leptospiral Uveitis

1317

Table 3. Demographic characteristics of uveitis patients compared with controls.

Characteristics
Age, years
Average age
Range
Sex
Male
Female
Place of residence
Rural
Urban
Animal-related occupations*
Socioeconomic class
Low
High

Uveitis patients
(n 46)

Uveitis controls
(n 49)

Cataract controls
(n 54)

Total

33
17 65

36
19 65

36
14 69

35
14 69

35 (74)
11 (26)

28 (57)
21 (43)

39 (72)
15 (28)

102 (68)
47 (32)

33 (72)
13 (28)
28 (63)

31 (63)
18 (37)
22 (45)

23 (44)
21 (56)
11 (20)

87 (58)
62 (42)
61 (41)

24 (52)
22 (48)

25 (51)
24 (49)

35 (67)
16 (33)

84 (56)
65 (44)

NOTE. Data are no. (%) unless stated otherwise.

* Includes farmers, cattle drivers, mutton-shop workers, rice sellers, goat herders, and pond clothes washers.

Rs 2000/month.

Table 4. Microagglutination test (MAT) titers by serovar for uveitis

patients.
MAT titers
Serovar
bulgarica
copenhagen
australis
hardjo
panama
medanenis
canicula
pomona
grippothyphosa
javanica
szwazijak
Total

1/50

1/100

1/200

1/400

6
6
5
3
3
2

3
1
1

1
1

1
1
1
28

1/800

1
1
1

Total
13
7
7
6
5
2
2
1
1
1
1
46*

NOTE. Data are no. of patients.

* 30% of seropositive uveitis patients reacted to 1 serovar; thus, 46 does
not reflect case no.

/ 9d45$$my18

03-03-98 14:07:02

jinfas

for leptospiral antibodies by MAT or ELISA (or both). Of the

PCR-positive uveitis and cataract controls, only 1 of the cataract controls was seroreactive, with an MAT titer of 1:50. Only
9 (19%) of 48 PCR-negative uveitis controls and 9 (18%) of
50 PCR-negative cataract controls had positive serology.
None of the individuals with prior evidence of leptospirosis
in either control group were positive by PCR except for 1
patient in the cataract group. For controls without prior leptospirosis, 1 (3%) of 39 uveitis controls and 3 (6%) of 48
cataract controls were PCR-positive. All 22 uveitis patients
with prior leptospirosis were PCR-positive. However, 13 (54%)
of 24 uveitis patients without evidence for prior leptospirosis
were also PCR-positive. The high rate of PCR positivity among
individuals with no prior evidence for leptospirosis may be
explained by the difficulty in diagnosing leptospirosis, the limitations in obtaining sera at the appropriate time intervals for
MAT testing, the limitations of the MAT assay itself, the fact
that some patients will be MAT-negative despite having had
systemic disease, and the fact that some patients may have had
subclinical and, therefore, undiagnosed leptospirosis who went
on to develop uveitis.
Data analyses for ocular characteristics. The ocular profile of uveitis patients is shown in comparison with uveitis
controls in table 6. Only 31 uveitis patients (67%) complained
of visual deficit compared to 46 uveitis controls (94%; P
.002). The majority of uveitis controls had anterior uveitis,
while the uveitis patients (most of whom were PCR-positive
[see above]) presented with panuveitis (P .001). Of the
panuveitis patients, 35% had bilateral disease. In the anterior
segment findings, only anterior chamber cells were significantly
associated with uveitis patients by multivariate analysis. In the
posterior segment findings, significant differences by multivari-

UC: J Infect

Downloaded from jid.oxfordjournals.org by guest on July 13, 2011

patients who were seropositive by MAT, 13 (54%) were also

IgM-positive. Two (25%) of 8 MAT-positive uveitis controls
were IgM-positive, as were 3 (43%) of 7 MAT-positive cataract
controls.
Results of PCR for leptospiral DNA. Leptospiral DNA was
detected in the aqueous humor of 37 uveitis patients (80%)
compared with 1 uveitis control (2%; P .001) and 4 cataract
controls (7%; P .001; table 5). Of 37 PCR-positive uveitis
patients, 24 (65%) were also seropositive for leptospiral antibodies by MAT or ELISA. Of the 9 uveitis patients who did
not demonstrate leptospiral DNA by PCR, 6 were seropositive

1318

Chu et al.

JID 1998;177 (May)

Table 5. Comparison of uveitis patients with uveitis and cataract controls by serology and polymerase
chain reaction (PCR).

Test
Serology/
IgM/
MAT 1:50
PCR/

Uveitis patients
(n 46)
33
22
24
37

Uveitis controls
(n 49)

(72)
(48)
(52)
(80)

10
8
8
1

Cataract controls
(n 54)

(20)
(16)
(16)
(2)

13
10
7
4

(24)
(19)
(13)
(7)

P*
.001

.001

NOTE. Data are no. (%).

* Statistical significance calculated as uveitis patients vs. both control groups; independent calculations were
equivalent.

Positive serology was defined as either IgM/ or MAT 1:50.

PCR was performed on aqueous humor only.

Discussion
In 1994, we found evidence to suggest that uveitis may occur
in epidemic form following outbreaks of leptospirosis in South
India [12]. Previously, specific diagnostic techniques were not
available to conclusively link this systemic infection with ocular disease. Culture is known to have a low sensitivity. MAT,
which detects serum antibodies to common leptospiral antigens,
is also limited in sensitivity and specificity [15]. PCR, however,
has become the most effective test for diagnosing leptospirosis
[16, 17]. This technique has recently been used to detect leptospiral DNA in the aqueous humor of 4 individuals with acute
anterior uveitis [18, 19]. We used this technique along with
MAT to show that Leptospira are associated with a high number of patients with uveitis among a population residing in a
leptospira-endemic region of South India.
Thirteen (28%) of PCR-positive uveitis patients did not demonstrate antileptospiral antibodies. Positive serology alone,
then, may not provide definitive evidence that uveitis was

/ 9d45$$my18

03-03-98 14:07:02

jinfas

caused by leptospira. Indeed, it remains unknown whether leptospiral uveitis correlates with serum antibody titers. The natural course of systemic titers varies; titers can remain elevated
for up to 2 years with levels as high as 1:16,400 in acute and
chronic disease [20] or decline within months following the
acute infection [21]. Many of these 13 uveitis patients developed uveitis up to 18 months after initial illness, which could
explain the low or absent titers. Furthermore, although Tamil
Nadu is an area with endemic leptospira, not all persons who
are exposed will seroconvert. This is supported by a study of
69 meat inspectors, of whom 12% who were diagnosed by
culture remained seronegative 2 years after the acute infection
[21]. Alternatively, not all individuals with titers to leptospira
would be expected to develop uveitis. In our study, 13 (24%)
of the cataract controls were seropositive and had no evidence
for uveitis. Many of these people were seen months after their
acute febrile episode and so presumably had sufficient time to
progress to ocular disease.
PCR provided additional data to determine which individuals
might have leptospiral uveitis. We found a statistically significant correlation between PCR results and uveitis patients compared with controls. Thirty-seven uveitis patients (80%) were
positive compared with 1 uveitis control (2%) and 4 cataract
controls (7%). It has been hypothesized that the organism may
release toxins or enzymes that directly induce a pathogenic
reaction. Evidence that a toxin is involved is supported by
subcellular damage that occurs in hepatocytes and renal epithelial cells [7] and the fact that many persons with leptospirosis
have an endotoxin-like shock syndrome [22].
Austoni et al. [23] has proposed that uveitis occurs because antibodies are slow to migrate into the anterior chamber but are rapidly cleared, allowing the organism to flourish.
In our study, 5 controls were PCR-positive yet did not develop uveitis. Thus, the presence of the organism or spirochetal load in the anterior chamber may not be sufficient
to induce an inflammatory response. Alternatively, these 5
controls may represent false-positive results for the assay.
Indeed, one PCR-positive uveitis control carried a diagnosis

UC: J Infect

Downloaded from jid.oxfordjournals.org by guest on July 13, 2011

ate analysis were vasculitis, vitreous opacities, and papillitis.

Figure 1 demonstrates the typical findings of vasculitis among
uveitis patients in this study.
We also compared the ocular profiles of all PCR-positive
individuals with those of all PCR-negative individuals (regardless of whether they were in the uveitis patient, uveitis control,
or cataract control group) and found results similar to those in
table 6; the same findings that were statistically significant
remained so. The ocular profile for MAT/ELISA-positive and
-negative individuals (again, regardless of group) were also
compared. The trends for statistical significance remained the
same, except that vitreous opacities were no longer significant.
Comparing the ocular profile for all PCR- and MAT/ELISApositive persons with all PCR- and MAT/ELISA-negative persons resulted in the exact same findings as in table 6.
The causes of uveitis among the controls are shown in table
1. All 27 individuals with idiopathic uveitis were negative by
serology for Leptospira organisms.

JID 1998;177 (May)

Leptospiral Uveitis

1319

of tuberculous uveitis. The clinical picture was quite distinct

from those described for the uveitis patients, and it is doubtful that the individual had 2 separate ocular infectious diseases. However, long-term follow-up of these controls would
be required to address this issue.
Nine of the uveitis patients were PCR-negative. Two of
the PCR-negative uveitis patients had posterior uveitis,
which may explain a negative PCR result for the aqueous
humor. This may be due to either insufficient assay sensitivity or absence of leptospira. Alternatively, immunologic
mechanisms without active infection could be responsible
for the remaining 7 uveitis patients who were PCR-negative.
Although we did not assay for leptospiral antibodies in the
aqueous humor, anterior chamber antibodies may occur coincident with systemic antibodies and have been described in
humans with leptospiral uveitis [24, 25]. Parma et al. [26]

Figure 1. Fundoscopic photograph from uveitis patient showing

prominent vasculitis.

No. (%)

Ocular characteristics

Uveitis patients
(n 46)

Ocular symptoms
Eye pain
Redness
Photophobia
Floaters
Defective vision
Location of uveitis
Anterior
Posterior
Panuveitis
Bilateral panuveitis
Anterior segment findings
Conjunctival injection
Keratic precipitates
Descemets membrane
folds
Anterior chamber flare
Anterior chamber cells
Hemorrhage
Hypopyon
Tubercles
Posterior synechiae
Posterior anterior synechiae
Busaccas nodules
Koeppes nodules
Posterior segment findings
Vasculitis
Vitreous opacities
Vitreous exudates
Chorioditis patch
Retinal exudates
Papillitis

Uveitis controls
(n 49)

31
27
32
22
31

(67)
(59)
(70)
(48)
(67)

38
23
37
26
46

(78)
(47)
(76)
(53)
(94)

NS
NS
NS
NS
.002

16
2
28
16

(35)
(4)
(61)
(35)

38
1
11
4

(78)
(2)
(22)
(8)

.001
NS
.002
.003

25 (54)
35 (76)

19 (39)
36 (73)

7
35
35
0
7
0
18
0
0
3

(15)
(76)
(76)

(7)

3
32
19
1
3
1
26
3
3
3

(6)
(65)
(39)
(2)
(6)
(2)
(53)
(6)
(6)
(6)

NS
NS
.001
NS
NS
NS
NS
NS
NS
NS

16
17
3
4
1
10

(35)
(37)
(7)
(9)
(2)
(22)

3
9
1
2
2
0

(6)
(18)
(2)
(4)
(4)

.001
.05
NS
NS
NS
.002

(15)
(39)

NS
NS

NOTE. Data are no. (%). NS not significant.

/ 9d45$$my18

03-03-98 14:07:02

jinfas

demonstrated that corneal epithelial cells cross-react with

leptospiral antigen and that no live organisms were needed
to induce antibodies to fix complement.
Leptospiral uveitis appears to be clinically distinct from
other uveitis entities. Studies conducted prior to 1960 have
classified leptospiral uveitis into two distinct categories: (1)
anterior uveitis with photophobia, blurred vision, and pain that
is self-limiting [8, 18] and (2) posterior segment involvement
or panuveitis with cotton-wool spots, chorioiditis, retinal hemorrhage, vitreous membranes, and papillitis [23, 27, 28]. Similar to the recent findings of Rathinam et al. [12], we found a
high prevalence of panuveitis among uveitis patients. Moreover, many of the clinical findings they presented could be
correlated with the PCR-positive uveitis patients in the present
study: anterior chamber cells, vitreous opacities, and vasculitis.
Vasculitis had not previously been reported in leptospiral uveitis but was observed among 16% of the uveitis patients in our
study. This finding is similar to the vasculitis observed in renal
vessels of individuals with systemic disease [11]. The pathogenic mechanism for this is unclear. Dobrina et al. [29] has
demonstrated that leptospira activate complement and stimulate
leukocyte phagocytosis, lymphocyte mitogenesis, and polymorphonuclear cell adherence. This may contribute to the development of vasculitis, whereby inflammatory cells are able
to migrate through leaky vessels and accelerate disease progression. Yet, it is unclear whether this occurs after the organisms
have been cleared from the eye or whether persistence of infection is responsible. The latter is supported by the large number
of uveitis patients in this study who had demonstrable leptospi-

UC: J Infect

Downloaded from jid.oxfordjournals.org by guest on July 13, 2011

Table 6. Ocular symptoms and signs for uveitis patients compared

with uveitis controls.

1320

Chu et al.

/ 9d45$$my18

03-03-98 14:07:02

jinfas

future research on the molecular epidemiology and immunopathogenesis of leptospira-induced uveitis.

Acknowledgments

We acknowledge the excellent technical assistance of Sister

Shantheeswari and the operating theater nurses and assistants in
the uvea clinic, Celine George of the Aravind Microbiology Laboratory, and Lee Smythe of the World Health Organization Leptospira Reference Laboratory, Brisbane, Australia.

References
1. Letocart M, Baranton G, Perolat P. Rapid identification of pathogenic
Leptospira species (Leptospira interrogans, L. borgpetersenii, and L.
kirschneri) with species-specific DNA probes produced by arbitrarily
primed PCR. J Clin Microbiol 1997; 35:248 53.
2. Johnson RC. Leptospirosis. In: Strickland GT, ed. Hunters tropical and
geographic medicine. Philadelphia: Saunders, 1990:889 93.
3. Kreiberg RA. An abundance of options. N Engl J Med 1993; 329:413 6.
4. Sehgal SC, Murhekar MV, Sugunan AP. Outbreak of leptospirosis with
pulmonary involvement in North Andaman. Ind J Med Res 1995; 102:
9 12.
5. Zaki SR, Shieh WJ. Leptospirosis associated with outbreak of acute febrile
illness and pulmonary haemorrhage, Nicaragua, 1995. Lancet 1995; 347:
535 6.
6. ONeil KM, Rickman LS, Lazarus AA. Pulmonary manifestations of leptospirosis. Rev Infect Dis 1991; 13:705 9.
7. Farr RW. Leptospirosis. Clin Infect Dis 1995; 21:1 8.
8. Barklay S, Garzozi H. Leptospirosis and uveitis. Ann Ophthalmol 1984;
16:164 8.
9. Weil A. Uber eine Eigentumliche mit Tumor, Icterus, und Nephritis einhergehende akute Infektionskrankheit (English summary). Dtsch Arch Klin
Med 1886; 39:209 32.
10. Shpilberg O, Shaked Y, Maier MK, Samra D, Samra Y. Long term follow
up after Leptospirosis. So Med J 1990; 83:405 7.
11. Farrar WE. Leptospira species (leptospirosis). In: Mandel GL, Bennett JE,
Dolin R, eds. Principles and practices of infectious diseases. New York:
Churchill Livingstone, 1995:2137 41.
12. Rathinam R, Ratnam S, Selvaraj S, Dean D, Nozik R, Namperumalsamy
P. Uveitis associated with an outbreak of leptospirosis. Am J Ophthalmol 1997; 124:71 9.
13. Terpstra WJ, Ligthart GS, Schoone GJ. ELISA for the detection of specific
IgM and IgG in human leptospirosis. J Gen Microbiol 1985; 131:
377 85.
14. Dean D, Oudens E, Bolan G, Padian N, Schachter J. Major outer membrane
protein variants of Chlamydia trachomatis are associated with severe
upper genital tract infections and histopathology in San Francisco. J
Infect Dis 1995; 172:1013 22.
15. Gravekamp C, Van de Kemp H, Franzen M, et al. Detection of seven
species of pathogenic leptospires by PCR using two sets of primers. J
Gen Microbiol 1993; 139:1691 700.
16. Bal AE, Gravekamp C, Hartskeerl RA, De Meza-Brewster J, Korver H,
Terpstra WJ. Detection of leptospires in urine by PCR for early diagnosis of leptospirosis. J Clin Microbiol 1994; 32:1894 8.
17. Brown PD, Gravekamp C, Carrington DG, et al. Evaluation of the polymerase chain reaction for early diagnosis of leptospirosis. J Med Microbiol
1995; 43:110 4.
18. Merien F, Perolat P, Mancel E, Persan D, Baranton G. Detection of Leptospira DNA by polymerase chain reaction in aqueous humor of a patient
with unilateral uveitis. J Infect Dis 1993; 168:1335 6.

UC: J Infect

Downloaded from jid.oxfordjournals.org by guest on July 13, 2011

ral DNA. Our data suggest that leptospira may directly induce a
pathogenic response in the eye during active uveitis. However,
further studies are required to definitively determine the pathogenic mechanism of disease.
Bilateral panuveitis, papillitis, and lack of visual deficit were
also significantly associated with uveitis. Thus, the combination
of these six ocular findings (anterior chamber cells, vitreous
opacities, papillitis, vasculitis, bilateral panuveitis, and lack of
visual deficit) could be used in a clinical diagnostic algorithm
for leptospiral uveitis in leptospira-endemic regions where confirmatory serology and PCR are not available. Further, these
ocular findings can be used to differentiate ocular leptospirosis
from other similar uveitis entities, such as those caused by
Mycobacterium tuberculosis (TB). TB uveitis is a significant
cause of morbidity in regions where leptospirosis is also endemic and presents with some eye findings similar to those of
panuveitis. However, the overall ocular profile for TB is different, with choroidal tubercles, granulomas, and mutton-fat keratic precipitates [30] being more characteristic than the anterior chamber cells, vitreous opacities, papillitis, and vasculitis
found in the ocular leptospirosis uveitis patients in this study.
Although more research is required, differentiating ocular leptospirosis from ocular TB will be important for determining
treatment regimens.
Currently, about half of the cases of uveitis are considered
idiopathic and most of these persons receive topical and systemic corticosteroids, which are inappropriate for treating most
individuals with infectious uveitis. We now show that a number
of these idiopathic cases of uveitis may be due leptospira.
During this study, almost a tenth of all persons seen in the
Uveitis Clinic at Aravind Eye Hospital were diagnosed with
leptospiral uveitis using MAT, PCR, and the clinical algorithm
that we describe here. In fact, we identified 46 uveitis patients,
the largest cluster of leptospiral uveitis patients reported so far.
This suggests that leptospira may be a significant cause of
uveitis, which until now has been largely unrecognized. Thus,
identifying leptospira as one of the causes of idiopathic uveitis
will dramatically affect treatment. Appropriate treatment, then,
may have an impact on morbidity by preventing recurrent disease and improving visual outcome.
In conclusion, Leptospira species have emerged as a significant cause of uveitis in South India. We show that both MAT
and PCR are important diagnostic tests and molecular epidemiologic tools for studying this disease. Leptospiral uveitis appears to be clinically distinct from other forms of uveitis; the six
ocular characteristics we describe above provide a diagnostic
profile that may impact treatment and visual outcome. Leptospira species should, therefore, be considered in the differential
diagnosis of uveitis among persons residing in or traveling to
Leptospira-endemic areas. We also provide molecular evidence
that leptospiral organisms are frequently found in the anterior
chamber during active uveitis, which may contribute to the
pathogenesis of this disease. This will be an important area for

JID 1998;177 (May)

JID 1998;177 (May)

Leptospiral Uveitis

19. Merien F, Baranton G, Perolat P. Comparison of polymerase chain reaction

with microagglutination test and culture for diagnosis of leptospirosis.
J Infect Dis 1995; 172:281 5.
20. Adler B, Faine S. The antibodies involved in the human immune response
to leptospiral infection. J Med Microbiol 1978; 11:387 400.
21. Blackmore DK, Schollum LM, Moriarty KM. The magnitude and duration
of titres of leptospiral agglutinins in human sera. NZ Med J 1984; 97:
83 6.
22. Watt G. Leptospirosis. In: Strickland GT, ed. Hunters tropical medicine.
Philadelphia: Saunders, 1991:317 23.
23. Austoni M, Moro F. Luveite quale manifestazione della sindrome oculoinfiammatoria della leptospirosi, con particolare riguarelo ad una forma
di ciclite sierosa silente in pazienti clinicalmente guariti. Giornale Mal
Infett Parassit 1952; 4:339; summary in Exerpta Med [Sect. 12] Ophthalmol 1954; 8:353.
24. Gsell O. Leptospirosis und Auge. Klin Monatsbl Augenheilkd 1952;
120:449.

1321

25. Doret M, Rohm A. Augenkomplikationen bei benignen Leptospirosen.

Klin Monatsbl Augenheilkd 1951; 118:51.
26. Parma AE, Santisteban CG, Villalba JS, Bowden RA. Experimental demonstration of an antigenic relationship between Leptospira and equine
cornea. Vet Immunol Immunopathol 1985; 10:215 24.
27. Levin N, Ngyuyen-Khoa JL, Charpentier D, Storbel M, Fournie-Amazouz
E, Denis P. Panuveitis with papillitis in leptospirosis. Am J Ophthalmol
1994; 117:118 9.
28. Duke-Elder S, Perkins ES. Diseases of the uveal tract. In: Duke-Elder S,
ed. Systems of ophthalmology. London: Kimpton, 1966:322 5.
29. Dobrina A, Nardon E, Vecile E, Cinco M, Patriarca P. Leptospira icterohemorrhagiae and Leptospire peptidoglycans induce endothelial cell
adhesiveness for polymorphonuclear leukocytes. Infect Immun 1995;
63:2995 9.
30. Dunn JP, Helm CJ, Davidson PT. Tuberculosis. In: Pepose J, Holland G,
Wilhelmus K, eds. Ocular infection and immunity. London: Mosby,
1996:1405 20.

Downloaded from jid.oxfordjournals.org by guest on July 13, 2011

/ 9d45$$my18

03-03-98 14:07:02

jinfas

UC: J Infect