Leptospirosis: a diagnostic conundrum

Introduction

Leptospirosis is a zoonosis, which is caused by spirochaetes belonging to the genus Leptospira. It has been reported to be endemic in several parts of South India such as Kerala, Tamil Nadu, Pondicherry and the Andamans. ¹

The clinical presentation varies from asymptomatic subclinical infection to severe Weil’s disease. ² The disease often goes unsuspected or misdiagnosed for want of laboratory resources, which are difficult to access and afford. Better diagnostics are needed from both clinical and public health points of view.

A variety of tests are available in the armament of the clinical microbiologist, but they need to be chosen prudently. Dark field microscopy is not widely available and often inaccurate. Culture, though important from an epidemiological perspective, is not of much use in routine clinical practice as it may take as long as three months to become positive.

Serology has been the mainstay of diagnosis. Microscopic agglutination test (MAT) is the widely acknowledged gold standard investigation though it is not available in all centres. IgM ELISA is a highly sensitive test which is widely available but is not sensitive enough to be used as the sole test.

Polymerase chain reaction (PCR) is a new generation molecular diagnostic modality with vast capabilities. Numerous sensitive, specific and rapid PCRs have been developed. LipL32 is an antigen present only in pathogenic Leptospira, which real-time PCR can detect earlier than other modalities. It also aids in quantitative monitoring which can be potentially used to monitor treatment efficacy. ³ Various protocols have been proposed which require further validation and optimisation. Our study was conducted to explore the usefulness of a highly sensitive and specific investigation with a short turnover time such as real-time PCR as a diagnostic modality for leptospirosis. We also aimed to detect the specific anti-leptospiral IgM antibody in the serum specimens of clinically suspected leptospirosis patients by ELISA.

Materials and methods

Our study was performed at a tertiary care teaching institute in Pondicherry from October 2013 to June 2015 after being approved by our Institute Research and Ethics Committees. The study included a single group of patients with possible but unconfirmed or presumptive diagnosis by Modified Faine’s criteria (Table 1).

OPEN IN VIEWER

Table 1.

Modified Faine’s criteria.

	Score
Part A: Clinical data
Headache	2
Fever	2
If fever, temperature ≥ 39℃	2
Conjunctival suffusion (bilateral)	4
Meningism	4
Muscle pain (especially calf muscle)	4
Conjunctival suffusion + meningism + muscle pain	10
Jaundice	1
Albuminuria or nitrogen retention	2
Part B: Epidemiological factors
Rainfall	5
Contact with contaminated environment	4
Animal contact	1
Part C: Bacteriological and lab findings
Isolation of Leptospira in culture – diagnosis certain
Positive serology
ELISA IgM positive*	15
SAT positive	15
MAT – Single high titre*	15
MAT – Rising titre (paired sera)	25

Patients with other established causes of prolonged fever were excluded.

Written informed consent was obtained from each patient or their parent/guardian and blood samples were collected before the administration of antibiotics.

The following tests were performed:

Culture: organisms were isolated from blood samples using EMJH medium;

From each patient, a blood sample was collected into two sterile vials, one being EDTA-coated for real-time PCR. Serum samples were stored in clean, sterile vials for IgM ELISA at −20℃. A few drops of blood were inoculated in two vials containing EMJH medium for culture of Leptospira according to the described method (WHO Leptospirosis - Laboratory Manual). ⁴

Duplicate samples were collected from each patient in EMJH medium, as described above, and were incubated in the dark at room temperature for a duration of six weeks. Cultures were examined for positive growth under a dark field microscope, on days 1, 3 and 5 followed by a 7–10-day interval up to a duration of six weeks, after which no growth was considered as negative. ⁴

Positive cultures were further confirmed by LipL32 real-time PCR.

Sera collected from blood samples were stored in clean vials at −20℃ for IgM ELISA testing. The assay used was Leptospira IgM Micro-well Serum ELISA (SCIMEDX Corp., Denville, NJ, USA). The assay was performed and interpreted according to the manufacturer’s instructions.

Blood samples collected in EDTA vials were stored at −20℃ for real-time PCR testing. The DNA was extracted according to the manufacturer’s instructions (QIAamp® Blood Mini Kit). Real-time PCR was performed to detect LipL32.

A case was considered positive based on Modified Faine’s criteria score (A+B+C) ≥ 25 (Table 1) and/or Leptospira DNA detected by PCR (WHO LERG report).

Results

Our study included 155 samples based on the inclusion criteria. Individuals hailed from Pondicherry and nearby regions of Tamil Nadu. Their mean age was 34.6 years (SD 13.7). The majority of patients were aged 26–45 years. Most of the study participants were men (103/155). Most of the suspected cases presented during the months of October to January.

Out of the 155 suspected cases, 14 were positive for IgM ELISA. Equivocal results were found in 29. The remaining 112 were negative. LipL32 real-time PCR was positive in 9/155 suspected cases. Three cases were culture-positive which was confirmed by LipL32 real-time PCR. Two cases were positive for both IgM ELISA and LipL32 real-time PCR. Overall, 21/155 were confirmed to be leptospirosis according to the case definition mentioned above. The positive test results are presented in Table 2.

OPEN IN VIEWER

Table 2.

Positive test results.

Positive test results	Cases (n)
IgM ELISA only	12
Real-time PCR only	4
IgM ELISA + real-time PCR	2
Real-time PCR + Culture	3

The majority of suspected cases presented in the second and third week of illness. The duration of illness at the time of investigation among confirmed cases of leptospirosis is presented in Table 3.

OPEN IN VIEWER

Table 3.

Correlation of days of illness (DOI) at the time of investigation with positive test results.

Days of illness (DOI)	IgM ELISA	Real-time PCR	Culture
≤ 7	0	6	3
8–14	8	3	0
15–20	5	0	0
≥ 21	1	0	0

Fever and headache were the commonest presenting features (100%) followed by myalgia (77%) and conjunctival suffusion (71%). The median Faine’s score was 21 in both suspected and confirmed cases.

Discussion

Leptospirosis is a septicaemic zoonosis present worldwide but more common in the tropical countries. Within India, it is more prevalent in places with high average rainfall and high soil water-holding capacity. The Andaman Islands are among the most endemic areas in the world with an incidence rate of 50/100,000. ⁵ However, an increasing incidence in Kerala, Tamil Nadu and Puducherry has been noted during the last two decades, including in the Andamans. This may be due to increased farming activity and inadequate rodent control. ⁶

In our study, 14% (21/155) of suspected cases fit the diagnosis of leptospirosis by case definition. The reason for a low positivity rate among suspected cases may be due to the inaptness of modified Faine’s criteria as an inclusion criterion and the non-specific clinical spectrum of the disease.

Isolation of Leptospira by culture is a definite proof of infection. Many reports have even suggested that culture is superior to MAT as a reference test.^7–9 However, culture is not of much clinical use as its sensitivity is low ¹⁰ and takes too long to become positive. It definitely has an important role in determining the global epidemiology of infection ⁷ and can help public health authorities in developing strategies for prevention.^7,10 Culture also provides important material for research. ⁷ In our study, culture was positive only in 3/21 confirmed cases. It also has to be noted that culture positivity was found only in cases that presented during the first week of illness. All culture positive cases were also positive for LipL32 real-time PCR. In a Thai study, culture positivity was only 29%. ¹¹ Isolation of leptospira by culture from blood samples was more successful compared to tissue samples. ¹²

Traditionally MAT has been considered to be the gold standard investigation.

However, it is not performed routinely in all laboratories owing to practical difficulties. It also has low sensitivity in acute phase sera. ⁸

IgM ELISA has been considered a method of choice for rapid diagnosis of leptospirosis. ¹⁰ The World Health Organization (WHO) recommends IgM ELISA for the sero-diagnosis of leptospirosis where healthcare resources are limited, although its reported accuracy is variable. As compared to culture, MAT and PCR, IgM ELISA is inexpensive and simple to perform. The results are objective and reproducible. ¹³ Many studies have demonstrated that ELISA is capable of detecting antibodies earlier than MAT.^10,13,14 In our study, IgM ELISA was positive in 14/21 confirmed cases in this study, but none presenting during the first week of illness were positive. Previous seropositivity rate was found to be only 23%. ¹⁵

PCR can detect circulating DNA long before antibodies were detected by serology. ¹⁶ Hence, its role in the diagnosis of the acute phase of illness is undisputed. LipL32 is present only in pathogenic leptospira. ¹⁷ We preferred LipL32 real-time PCR as it can be performed quickly (3 h) compared to the conventional PCR method. ¹⁸

Compared to our study, where LipL32 real-time PCR was found to be positive in 9/21, a Uruguayan study ¹⁹ showed the diagnostic sensitivity of PCR in comparison with MAT to be only 30%. Many other investigators have reported values close to 50%. However, the specificity was 100%. Low diagnostic sensitivity may be due to low bacteraemia in mild cases, collection of samples many days after the onset of symptoms especially when antibodies are present or even after starting antibiotic therapy.^11,20 Samples stored at −20℃ can result in a decrease in the number of leptospira. ¹⁹

The window of PCR and culture positivity is within one week of developing clinical manifestations.^10,11 The window period for PCR is slightly longer than that of culture. We found that real-time PCR was useful for diagnosis in those cases that presented during the first 10 days of illness. Real-time PCR has some drawbacks too. One major factor preventing routine implementation is the cost, which is high in terms of equipment and training. ¹¹

Limitations to our study are that we investigated patients consulting at hospital and may have missed other cases in the community with milder manifestation. Thus, the prevalence of leptospirosis in this region cannot be assessed. Hospitalised patients are a priori a population sicker than ambulatory patients and thus represent a biased sampling. It is also possible that many mild cases might have been treated at lower rung hospitals and were not referred. Generalisation of our results should be made with caution.

Conclusion

In our opinion, it is high time the diagnostic approach to leptospirosis is changed. Real-time PCR must be considered for all suspected cases. Provision should be made for regional laboratories in endemic areas and made accessible to primary care physicians.