Utility of Blood as the Clinical Specimen for the Molecular Diagnosis of Post-Kala-Azar Dermal Leishmaniasis

ABSTRACT

The countries in the Indian subcontinent have reported a dramatic decline in visceral leishmaniasis (VL) cases. However, the presence of the parasite reservoir in the form of post-kala-azar dermal leishmaniasis (PKDL), a dermal sequel of VL, is a hurdle in attaining VL elimination. Presently employed clinical specimens for the diagnosis of PKDL include skin biopsy specimens and slit skin smears. In this study, the use of blood as a clinical specimen was investigated in different manifestations of PKDL in India. This is a bicentric study (National Institute of Pathology, Indian Council of Medical Research [ICMR], New Delhi, and Institute of Medical Sciences [IMS], Banaras Hindu University, Varanasi), with 215 participants (120 PKDL patients and 95 controls). Highly sensitive quantitative real-time PCR (Q-PCR) and field-deployable loop-mediated isothermal amplification (LAMP) were employed using blood samples for diagnosis. Promising sensitivities of 77.50% (95% confidence interval [CI], 69.24 to 84.05%) for Q-PCR and 70.83% (95% CI, 62.16 to 78.22%) for LAMP were obtained for the diagnosis of PKDL. Further, enhanced sensitivities of 83.33% (95% CI, 71.28 to 90.98%) and 77.78% (95% CI, 65.06 to 86.80%) for Q-PCR and LAMP, respectively, were recorded for the detection of macular cases. The study revealed an inverse correlation between the parasite load estimated in slit and blood samples, thereby favoring the use of blood for the diagnosis of the macular variant, which may be missed due to scant parasite loads in the slit. This study is the first to propose the promising potential of blood as a clinical specimen for accurate diagnosis of PKDL, which would aid in fast-tracking VL elimination.

INTRODUCTION

Leishmaniasis is a vector-borne infectious disease, caused by the genus Leishmania. As reported by the World Health Organization (https://www.who.int/news-room/fact-sheets/detail/leishmaniasis), it is among the deadliest neglected tropical diseases, afflicting nearly 700,000 to 1 million people annually. Visceral leishmaniasis (VL), also known as kala-azar (KA), is the disease's severest form. Post-kala-azar dermal leishmaniasis (PKDL), a rare dermatosis, is reported in 2.5 to 20% of patients who have recovered from VL in Asia (1) and in 50 to 60% cases in Africa (2). Approximately 10 to 23% of PKDL patients report no history of previous VL infection (1, 3, 4). The clinical presentation of the disease primarily includes polymorphic lesions with co-occurrence of macules and papulonodules (5). Unlike the African PKDL, a grading system has not been defined for classification of Indian PKDL. The clinical presentation includes (i) macular PKDL, predominantly consisting of hypopigmented or hypomelanotic lesions; (ii) papulonodular PKDL, characterized by papules and/or nodules; and (iii) mixed or polymorphic PKDL, where patients present with all three types of lesions, i.e., macular, papular, and nodular (4). The presence of PKDL perpetuating the anthroponotic transmission of Leishmania donovani in the Indian subcontinent (ISC) is jeopardizing the ongoing VL elimination effort. The initial WHO neglected tropical disease (NTD) road map (2012 to 2020) aimed at elimination of VL as a public health problem by 2020. Now, the WHO NTD road map has proposed a new target of achieving a <1% rate of case fatalities due to primary VL. Furthermore, it includes a subtarget for validating the elimination, i.e., <1 case (new and relapses) per 10,000 population at the block level in India at 100% by 2030 (6). (A block is a district subdivision and represents a compact area for which effective plans are prepared and implemented through village councils in rural areas of India.)

Management of PKDL is challenging due to the lack of a definitive diagnostic test. The primary diagnosis of PKDL still relies heavily on the combination of clinicohistopathological examination, rK39 antibody test, history of VL, and exclusion of associated dermatosis requiring differential diagnosis. The conventional gold standard remains the demonstration of parasites in skin biopsy specimens or slit skin smears (SSS); however, its limitations are the need for invasive procedures, the requirement for technical expertise, and an unacceptably low detection range of 4 to 58% (7, 8). The rK39 rapid diagnostic test (RDT) is widely employed for detection of both VL and PKDL. However, the rK39 RDT cannot be used as a confirmatory diagnostic tool for PKDL, due to the presence of past antileishmanial antibodies. Molecular diagnostics overcome these constraints with high sensitivity and specificity, along with applicability for different clinical specimens from PKDL patients. The sensitivity of conventional PCR, nested PCR, and quantitative PCR (Q-PCR) with different targets and clinical specimen ranges from 76 to 100%, while the specificity was 100% in all assays (9). However, these molecular methods face logistic issues, which act as a stumbling block in their field application. Recently, loop-mediated isothermal amplification (LAMP) has shown promising results in the field of diagnostics by amplifying DNA with high sensitivity and specificity under isothermal conditions, without the requirement of any sophisticated instrumentation. It has been widely applied for the diagnosis of leishmaniasis, with sensitivity and specificity ranging between 80 and 100% and 94 to 100%, respectively (10). We have successfully established and validated a SYBR green I-based closed-tube LAMP assay with an excellent sensitivity of 97% and specificity of 100% using blood and bone marrow aspirates for diagnosis of VL and tissue biopsy specimens for PKDL (11, 12). Additionally, direct blood lysis (DBL)-LAMP, which eliminates the need for DNA isolation, has also been reported for detection of VL (12).

The clinical specimens employed so far for PKDL include skin biopsy specimens and SSS. The aim is to employ a patient-friendly, minimally invasive, technically less demanding, and field-applicable clinical sample. Considering the above and the high rate of PKDL patients’ refusal to provide skin biopsy specimens, the applicability of using SSS for Q-PCR has already been established for diagnosis and the assessment of cure in PKDL (13, 14). Moreover, with a threshold limit of detection of 4 parasites/μl from slit aspirate, it has proven to be more sensitive than tissue biopsy specimen microscopy (15). The utility of blood in Q-PCR for VL diagnosis is well established, with a reported optimized sensitivity of up to 0.0125 parasite/ml in blood (16). However, the utility of blood as a clinical sample for diagnosis of PKDL has not been investigated so far.

The present study, to our knowledge, is the first to report positivity of blood using Q-PCR and LAMP for the diagnosis of PKDL in a large number of cases. The results obtained by us correlated with the results on the PKDL samples from the other center of the study. Detection of PKDL using blood has shown more promising results in the macular variants than the polymorphic cases. The study is of paramount importance in terms of applicability of the hitherto-unexplored use of blood specimens in highly sensitive Q-PCR and field-deployable LAMP for accurate and rapid detection of “parasite carriage” in the form of PKDL.

MATERIALS AND METHODS

Study design.

This was a long-term (2015 to 2019), bicentric study conducted at the following locations. The first is the National Institute of Pathology (NIP), Indian Council of Medical Research (ICMR), New Delhi, India (analysis center). The samples were collected at the Department of Dermatology, Safdarjung Hospital, New Delhi, India (collection center). We refer to these two centers together as site 1 here. The second is the Institute of Medical Sciences (IMS), Banaras Hindu University (BHU), Varanasi, India (analysis center). The samples were collected at their field site, Kala-Azar Medical Research Centre (KMRC), Muzaffarpur, Bihar, India (collection center). We refer to these two centers together as site 2 here.

The study population included 215 participants comprising 160 individuals at site 1 and 55 individuals at site 2.

(i) Patient group.

A total of 120 PKDL patients were included in the patient group, comprising 54 macular and 66 polymorphic cases. At site 1, 80 patients (31 with macular and 49 with polymorphic PKDL) diagnosed with PKDL on the basis of clinicohistopathological presentation of macular/maculopapular or nodular rash were recruited. The cases were confirmed by microscopy and/or Q-PCR performed using slit aspirates. Similarly, 40 individuals (23 with macular and 17 with polymorphic PKDL) clinically diagnosed with PKDL reporting at site 2 were enrolled. The cases were confirmed by microscopy and/or PCR from tissue biopsy specimen. Individuals suffering from any coinfections, HIV infection, or other preexisting infectious conditions and pregnant/lactating women were excluded from the study.

(ii) Control group.

The control group comprised 95 individuals at both centers of the study. A total of 80 controls were included in study at site 1, which comprised 50 controls with other differential dermal diseases (leprosy [n = 30] and vitiligo [n = 20]) and 30 healthy controls (healthy controls from areas where PKDL is endemic [n = 15] and healthy controls from areas where PKDL is not endemic [n = 15]). A total of 15 healthy controls from areas where PKDL is not endemic were tested at site 2. The samples collected included blood and slit skin smears (SSS) from controls with other dermal diseases. Only blood was procured from healthy controls.

For recruitment of control samples, the nondiseased individuals who voluntarily participated at site 1 were included as healthy controls. Patients with other differential diseases were clinically confirmed cases, reporting at site 1. Leprosy cases were confirmed by microscopy and Q-PCR. Vitiligo cases were confirmed on the basis of clinical examination conducted by a dermatologist. Informed written consent was obtained from all participants for collection and subsequent analysis of the samples. After extraction of DNA from the collected samples, it was subjected to both Q-PCR and LAMP. Figure S1 in the supplemental material gives a schematic flow diagram of the study design.

Study ethics.

Patients’ recruitment was carried out in conformity with the principles of the Declaration of Helsinki 1975 on human rights, revised in Edinburgh, 2000. The PKDL study at site 1 obtained ethical clearance under the guidelines of the ethics committee of Safdarjung Hospital, Vardhman Mahavir Medical College, New Delhi, India (VMMC/SJH/PROJECT/221012/7). For PKDL cases enrolled at site 2, approval was granted from the institutional review board of Banaras Hindu University, Varanasi, India (Dean/2017/EC/185). Written informed consent was obtained from all participants for collection of samples and their further analysis.

DNA isolation from clinical samples.

Two hundred microliters of blood was collected from PKDL patients in heparinized tubes. DNA extraction was carried out from blood as per the instructions provided with QIAamp DNA blood minikit (Qiagen, Hilden, Germany). Slit skin aspirates (SSS) were collected in 200 μl of NET buffer (150 mmol/liter NaCl, 15 mmol/liter Tris-HCl [pH 8.30], and 1 mmol/liter EDTA) using a previously described slit scrape technique (13) under sterile conditions. At site 2, tissue biopsy specimens (3- to 4-mm punch biopsy specimen) were collected from PKDL cases in NET buffer. DNA extraction from tissue biopsy specimen and SSS was performed using the QIAamp DNA tissue kit (Qiagen, Hilden, Germany) as per the manufacturer’s protocol. DNA from blood, tissue, and slit aspirates was eluted in 50 μl, 100 μl, and 25 μl of nuclease-free water, respectively.

Molecular methods. (i) PCR and Q-PCR.

At site 1, SYBR green I-based Leishmania genus-specific real-time Q-PCR was performed for the confirmation and estimation of parasite load in PKDL patients. The primers were specific for the L. donovani kinetoplast minicircle DNA (kDNA) region. The reaction was carried out in an ABI Prism 7500 sequence detection system (Applied Biosystems, CA, USA) in triplicate, employing primers and conditions described previously (17). PKDL diagnosis at site 2 was confirmed by performing PCR using tissue biopsy samples with L. donovani species-specific primers. The forward and reverse primers, together designated LdI primers, were designed against L. donovani kDNA. An amplified product of approximately 600 bp was visualized on gels (18).

(ii) LAMP assay.

The LAMP assay (visual LAMP) was carried out in batches of 10 reactions/set, including a positive control (1 ng/μl parasite genomic DNA), a no-template control, a confirmed VL blood DNA sample, samples from patients with other diseases or healthy controls, and 6 test samples. Leishmania donovani AG83 (MHOM/IN/83/AG83) was included as a positive control. The established SYBR green I-based closed-tube LAMP assay was performed as described previously (11). A reaction of 25 μl was set up, consisting of 40 pmol each of forward inner primer (FIP) and backward inner primer (BIP), 5 pmol each of F3 and B3 primers, 20 pmol each of forward loop primer (FLP) and backward loop primer (BLP), 1.4 mM (each) deoxynucleoside triphosphate, 0.8 M betaine, 20 mM Tris-HCl, 10 mM (NH₄)₂SO₄, 8 mM MgSO₄, 0.1% Triton X-100, 8 U of Bst 2.0 warm-start DNA polymerase (New England Biolabs, Ipswich, MA, USA), and 2 μl of DNA sample. On the inner side of the tubes, 1 μl of 1:10 diluted SYBR green I (10,000× concentrate in dimethyl sulfoxide [DMSO]) (Molecular Probes, Eugene, OR, USA) was placed before the reaction was set up. Incubation was done at 65°C for 60 min. Postreaction, the tubes were given a brief spin for mixing of SYBR green I with the amplified product. An instantaneous color change to green was observed in the positive samples, while the negatives stayed orange/unchanged.

Statistical analysis.

Parasite loads are presented as medians (interquartile ranges). All statistical analyses were performed using GraphPad Prism software, version 5.01 (GraphPad Software, San Diego, CA). A Mann-Whitney test was performed to compare the two groups. A test was considered significant if the P value was less than 0.05. Receiver operating characteristic (ROC) curves were made to determine the diagnostic accuracy of Q-PCR and LAMP. Comparative ROC curves were drawn using IBM SPSS Statistics software V26 (SPSS, Chicago, IL, USA). Area under the curve (AUC_ROC) was computed using the trapezoidal rule. In general, an AUC_ROC of 0.5 suggests no discrimination (i.e., the ability to diagnose individuals with and without disease based on the test); a value of 0.7 to 0.8 is considered to indicate acceptable diagnostic accuracy, a value of 0.8 to 0.9 is considered excellent, and a value greater than 0.9 is considered outstanding (19).

RESULTS

Study population.

At site 1, a total of 80 PKDL patients included 55 males (68.75%; median age, 24 years [7 to 60 years]) and 25 females (31.25%; median age, 25 years [13 to 55 years]). A majority 76 of 80 (∼95%) patients reported a history of VL, while 4 patients reported no prior VL episode. Moreover, 44 of 76 cases (∼58%) cases manifested PKDL within 5 years after cure of VL. In terms of categorization on the basis of clinical manifestations, 21 of 30 (70%) patients with macular cases and 23 of 46 (50%) patients with polymorphic cases reported VL within 5 years following cure of VL.

At site 2, a total of 40 PKDL patients were enrolled, including 28 males (70%; median age, 32 years [11 to 66 years]) and 12 females (30%; median age, 34 years [14 to 59 years]). A total of 31 of 40 (∼77.5%) cases had had a previous episode of VL, while 9 patients reported no history of VL. Furthermore, 18 of 31 cases (58%) presented with PKDL within a period of ≤5 years following cure of VL. In terms of clinical categories, 12 of 19 (63.15%) patients with macular variants and 6 of 12 (50%) patients with polymorphic variants reported PKDL in ≤5 years after VL.

The control group comprised a total of 95 individuals at both sites. The group included 55 males (57.89%; median age, 29 years [8 to 65 years]) and 40 females (42.10%; median age, 23 years [4 to 60 years]).

Sensitivity and specificity of Q-PCR and LAMP for diagnosis of PKDL using blood DNA.

The results at the two sites are discussed individually and after combined analysis, as two different methods were used for confirmation of PKDL at the two sites. At site 1, Q-PCR using slit aspirates was performed to quantify parasite load in all cases, whereas at site 2, conventional PCR using tissue biopsy specimens was done, and thus, the parasite load was not quantified.

(i) Site 1 (NIP, ICMR).

All cases of PKDL were confirmed by microscopy and/or Q-PCR performed using slit aspirates. Q-PCR using slit aspirates was considered the reference standard. Parasite DNA was detected in 61 of 80 PKDL blood DNA samples by Q-PCR, giving it a sensitivity of 76.25% (95% confidence interval [CI], 65.86 to 84.24%); 55 of 80 cases were positive by the LAMP assay, giving it a sensitivity of 68.75% (95% CI, 57.93 to 77.85%). All 80 controls, including leprosy patients and healthy controls, were negative by both assays, indicating a specificity of 100% (95% CI, 95.42 to 100%).

(ii) Site 2 (IMS, BHU).

Q-PCR detected Leishmania DNA in 32 of 40 PKDL blood DNA samples, yielding a sensitivity of 80% (95% CI, 65.24 to 89.50%). Thirty of 40 cases were positive by visual LAMP, indicating a sensitivity of 75% (95% CI, 59.81 to 85.81%). Both Q-PCR and visual LAMP were negative for 14 of 15 controls tested, giving a specificity of 93.33% (95% CI, 70.81 to 98.81%).

Overall, including samples at both the centers, Q-PCR in the detection of PKDL using blood DNA was positive in 93 of 120 cases tested, yielding a sensitivity of 77.50% (95% CI, 69.24 to 84.05%). For LAMP, 85 of 120 samples were positive, giving a sensitivity of 70.83% (95% CI, 62.16 to 78.22%). Both assays showed negative results with 94 of 95 controls tested, providing an overall specificity of 98.95% (95% CI, 94.28 to 99.81%). The combined sensitivities obtained, based on clinical manifestation of the PKDL, were as follows. Q-PCR detected 45 of 54 macular cases, yielding a higher sensitivity of 83.33% (95% CI, 71.26 to 90.98%) than in detection of 48 of 66 polymorphic variants, indicating a sensitivity of 72.73% (95% CI, 60.96 to 82%). Similarly, LAMP successfully detected parasites in 42 of 54 macular cases, giving a sensitivity of 77.78% (95% CI, 65.06 to 86.80%), versus 43 of 66 polymorphic cases, indicating a sensitivity of 65.13% (95% CI, 53.11 to 75.52%) (Table 1).

TABLE 1.

Comparative sensitivity and specificity of Q-PCR and visual LAMP for diagnosis of post-kala-azar dermal leishmaniasis using blood DNA at the two sites

Site and molecular diagnostic test	Sensitivity in PKDL cases						Specificity in controls
	Macular		Polymorphic		Combined		Ratiob	% (95% CI)
	Ratioa	% (95% CI)	Ratioa	% (95% CI)	Ratioa	% (95% CI)
NIP, ICMR							80/80	100 (95.42–100)
Q-PCR	26/31	83.87 (67.37–92.91)	35/49	71.43 (49.27–75.33)	61/80	76.25 (65.86–84.24)
LAMP	24/31	77.42 (60.91–88.60)	31/49	63.27 (49.27–75.33)	55/80	68.75 (57.93–77.85)
IMS, BHU							14/15	93.33 (70.18–98.91)
Q-PCR	19/23	82.61 (62.86–93.02)	13/17	76.47 (52.74–90.44)	32/40	80 (65.24–89.50)
LAMP	18/23	78.26 (58.10–90.34)	12/17	70.59 (46.87–86.72)	30/40	75 (59.81–85.81)
Combined							94/95	98.95 (94.28–99.81)
Q-PCR	42/54	83.33 (71.26–90.98)	48/66	72.73 (60.96–82.0)	93/120	77.50 (69.24–84.05)
LAMP	42/54	77.78 (65.06–86.80)	43/66	65.15 (53.11–75.52)	85/120	70.83 (62.16–78.22)

^aTrue positives/(true positives + false negatives).

^bTrue negatives/total negatives.

Comparative ROC curve analysis to determine the diagnostic accuracy of Q-PCR and LAMP for diagnosis of PKDL.

Comparative ROC curves were drawn for comparing the diagnostic accuracy of Q-PCR and LAMP for diagnosis of PKDL. Both Q-PCR and LAMP demonstrated excellent diagnostic accuracy, with AUC_ROC of 0.882 and 0.849, respectively, for diagnosis of PKDL using blood (Fig. S2A). Comparison of diagnosis of different clinical presentations indicated that for macular PKDL, Q-PCR demonstrated outstanding diagnostic accuracy, with an AUC_ROC of 0.917. LAMP also exhibited excellent performance, with an AUC_ROC of 0.889 (Fig. S2B). For polymorphic PKDL, both Q-PCR and LAMP showed excellent diagnostic accuracy, with AUC_ROC of 0.864 and 0.826, respectively (Fig. S2C).

Comparative sensitivity and specificity of Q-PCR and LAMP for diagnosis of PKDL employing blood and slit samples (site 1).

The median parasite loads in slit aspirates of from patients with macular and polymorphic PKDL were 55 parasites/μl (range, 3 to 7,573/μl) and 856 parasites/μl (range, 4 to 573,946/μl) of slit aspirate, respectively. Significant differences in parasite load were evident between the two variants (P = 0.0020 and P < 0.005). However, in blood samples from macular cases, a higher parasite load of 375 parasites/ml (range, 0 to 38,336/ml) was seen, compared to 50 parasites/ml (range, 0 to 4,386/ml) of blood in polymorphic variants. Remarkable differences were noted between the two types (P = 0.0135 and P < 0.05) (Fig. 1). Individual casewise analysis, for both variants, indicated an inverse relationship between the parasite load in slit aspirates and blood (Table S1 and Fig. 2). All 80 cases were positive by Q-PCR using slit aspirates, giving it a sensitivity of 100% (95% CI, 95.42 to 100%). LAMP detected parasite DNA in 72 of 80 cases, exhibiting an exemplary sensitivity of 90% (95% CI, 81.49 to 94.85%). Based on different manifestations, LAMP was positive in 26 of 31 and 46 of 49 cases, giving it sensitivities of 83.87% (95% CI, 67.37 to 92.91%) and 93.88% (95% CI, 83.48 to 97.90%) for macular and polymorphic variants, respectively. Of 80 cases, 4 (3 macular and 1 polymorphic) were negative by all the tests under consideration (Q-PCR with blood DNA and LAMP with slit and blood DNA) and were positive only by the reference standard Q-PCR with slit DNA (39, 4, 4, and 5 parasites/μl). Table S1 summarizes comparative Q-PCR and LAMP results using blood and slit samples for diagnosis of PKDL.

FIG 1.

Parasite loads in slit aspirates and blood of patients with post-kala-azar dermal leishmaniasis (PKDL) patients as determined by quantitative real-time PCR (Q-PCR). (A) Median parasite loads in blood of patients with macular and polymorphic variants. (B) Median parasite loads in slit aspirates of patients with macular and polymorphic types.

FIG 2.

Inverse correlation between the parasite load estimated in slit aspirates and blood from post-kala-azar dermal leishmaniasis (PKDL) patients. The polymorphic variants can be seen clustered at top left, indicative of higher parasite loads in slit aspirates and lower parasite loads in blood. Similarly, the macular variants are aggregated toward the lower right side of the graph, indicating lower parasite loads in slit aspirates and higher parasite loads in blood.

DISCUSSION

PKDL is a baffling and socially stigmatizing disease with significant underreporting in the ISC. Reported prevalence is 1.1 cases per 10,000 population recently (20) and was initially 4.4 per 10,000 (21). At its advent, PKDL was not included directly in target elimination, which led to its being often neglected and missed during active case searches. However, the new WHO NTD road map for eliminating VL incorporates a subtarget for detecting PKDL cases (VL cases to be followed up for 3 years) and treating 100% by 2030 (6).

It is well recognized that PKDL cases hold grave significance for the possibility of setting off a new epidemic of VL. It has been reported that PKDL patients are reluctant to seek treatment, despite having the condition for years (2), and thus are infectious for a longer duration than patients with active VL. Recent xenodiagnosis studies have revealed that patients with active VL and PKDL can transmit L. donovani to the sandfly vector, underscoring the significance of early diagnosis and treatment in obliterating these infection reservoirs (22). However, there remains a considerable gap in the field of diagnostics for PKDL, posing an obstacle to their management. PKDL is often confused with pauci- or multibacillary leprosy. Approximately 27% of cases of macular PKDL were misdiagnosed at the primary health center, of which 78% were misidentified as leprosy cases and were given partial or complete treatment for the same (4). Further, recent studies employing active case surveillance have revealed that macular PKDL constitutes almost 50% of the PKDL burden (23). However, no tool is available for accurate diagnosis of macular PKDL, which requires differential diagnosis. Standard microscopy also has an extremely low (7 to 33%) sensitivity of detection (24). Though macular cases have lower parasite loads in tissue than polymorphic cases, they play a definitive role in disease transmission (25).

Crafting an effective diagnostic algorithm for macular PKDL is the need of the hour. We reported earlier the utility of slit samples for the diagnosis of PKDL using both Q-PCR (13) and LAMP (26). The present study probed the potential application of blood as a clinical specimen for the diagnosis of PKDL (especially for the macular variant) employing Q-PCR and LAMP. Encouraging sensitivities of 77.50% (Q-PCR) and 70.83% (visual LAMP) were observed when blood was used for PKDL diagnosis. Moreover, higher sensitivities of 83.33% (Q-PCR) and ∼78% (LAMP) were observed for the detection of macular PKDL. Overall, blood as a clinical specimen yielded high sensitivity to detect PKDL, when either molecular diagnostic tool was used. Thus, we propose a revised clinical algorithm for confirmatory diagnosis of PKDL, based on the existing WHO algorithm (27) (Fig. 3). It includes the collection of blood as the first clinical specimen for diagnosis of probable PKDL, which is expected to yield a sensitivity ranging between 70 and 80%, with a provision of employing LAMP in low-resource settings. The negative cases may be further tested at referral centers with minimally invasive SSS, using microscopy and/or molecular diagnosis with Q-PCR. This new algorithm, if successfully validated under field conditions, will assist in rapid and reliable diagnosis of PKDL.

FIG 3.

Proposed clinical algorithm for rapid and confirmed molecular diagnosis of post-kala-azar dermal leishmaniasis (PKDL) in the Indian subcontinent. The revised algorithm employs both blood and SSS as clinical specimens for detection of PKDL.

PKDL is an immunologically triggered manifestation, with both the host and the parasite being involved in pathogenesis. PKDL is characterized by an intermediate position between T_H1 and T_H2 overlapping and counterregulatory responses, which translates into typical clinical features. The etiopathogenesis of PKDL is controversial. It is still unexplained and intriguing what drives the viscerotropic L. donovani causing VL to become dermatotropic, manifesting as PKDL. Theories supporting persistence and reinfection have both been proposed (28). Previous studies distinguishing macular and polymorphic immune responses have reported strong immunological response in macular PKDL, with heightened cell-mediated immunity (CMI), leading to lower parasite load in slit/tissue. Further, weaker humoral immunity with low antibody levels (only IgG1 is elevated) is reported in macular cases (29, 30). On the other hand, CMI is low in polymorphic variants, which is induced by transforming growth factor β (TGF-β) and interleukin 10 (IL-10), along with increased levels of markers for regulatory T cells (Tregs) (31–33), leading to a high parasite load in SSS or tissue. Moreover, a pronounced humoral response is seen in polymorphic cases, demonstrated by high antibody levels (IgG1 and IgG3) (30).

In our study, both Q-PCR and LAMP gave high positivity for detection of PKDL when blood was the clinical sample. An inverse correlation was observed between the parasite load in blood and slit aspirates. In polymorphic cases, the estimated parasite load was high in the slit aspirate and relatively low or nil in blood, whereas in macular cases, low parasite loads were seen in the slit and higher ones in blood. The macular variants usually have low/scant parasite loads in slit or tissue, in comparison to polymorphic cases (13, 17), as seen in the present study. However, our results when blood was used demonstrated a higher degree of positivity in the blood of patients with macular variants than in polymorphic variants. This suggestive inverse correlation has led us to propound that in macular variants, the parasite might be considered to be in the process of migrating from viscera to the dermis, thus showing parasite persistence in the systemic circulation. However, in coalesced dermal papular and nodular manifestations, the parasite has probably relocalized or limited itself to the dermis, thus showing lower loads in or complete absence from visceral circulation.

Overall, the immunopathogenesis of PKDL remains abstruse, and more in-depth studies are warranted for defining the role and interplay of systemic and lesional immune response in PKDL, supported by clinical evidence, to provide solid direction in delineating the much-debated conundrum of VL-to-PKDL transition.

Studies of the diagnosis of PKDL employing blood DNA are lacking. However, PCR using blood buffy coat has been reported and has shown a positive relationship between PKDL grade and PCR positivity. Approximately 65% of patients with grade II disease (maculopapular or nodular rash appearing mostly on the face and to the chest, back, upper arms, and legs) were positive by peripheral buffy coat PCR, compared to 12.5% of patients with grade I disease (scattered maculopapular or nodular rash on the face with or without some involvement of the upper chest or arms). These findings had led the investigators to suggest that PKDL patients may serve as the reservoir for the transmission of the LD parasites, which circulate in their blood (7).

Despite attempts to make this a comprehensive study, we encountered the following limitations. (i) The 1 healthy control from an area where PKDL is not endemic, out of 15 tested at site 2, who was marginally positive by both Q-PCR and LAMP was lost to follow-up. (ii) Data for parasite load in tissue samples of a few unpaired cases of PKDL at site 2 were missing.

Fundamentally, this study has unveiled and highlighted the practicality of blood as a clinical specimen for detection of PKDL, especially for the macular variants, by employing highly sensitive molecular tools, such as Q-PCR and field-friendly LAMP.