Abstract
Leprosy, caused by Mycobacterium leprae, is a devastating human disease where damage to the peripheral nerves leads to deformities of the hand, foot, face and in some cases eye. A strategy to diagnose leprosy at an early stage and subsequent treatment could help in preventing both serious disability in the patient and transmission, as untreated leprosy is contagious. As of now, accurate and efficient diagnosis of leprosy is considered as a great challenge for leprosy health care. Employing the skin smear test, leprosy patients are classified as smear-positive (for acid-fast bacilli) multibacillary (MB) and smear-negative (for acid-fast bacilli) paucibacillary (PB) types (World Health Organization, 1988). Regarding serology in leprosy, phenolic glycolipid-I (PGL-I) is a highly specific antigen of M. leprae (Hunter & Brennan, 1981) and is known to give rise, predominantly, to IgM antibodies against its terminal trisaccharide. Using PGL-I based antibody detecting tests, well over 90 % of MB patients have high levels of antibodies whereas 1540 % of the PB patients have detectable antibodies (Buhrer et al., 1998; Chanteau et al., 1989, 1993; Cho et al., 1983, 1991; Fujiwara et al., 1984; Hussain et al., 1990; Roche et al., 1991; Saad et al., 1990). Thus, the major limitation of the serological assays has remained the detection of a limited number of smear-negative PB patients that could be due to an undetectable weak antibody response to M. leprae. Hence, a strategy to improve the detection rate for such patients is clearly needed. In the present study we have investigated the effect of incubating the reactants at low temperature in order to enhance the performance for detecting antibodies to phenolic glycolipid-I antigen.
Study sera included samples from 56 leprosy patients collected at National JALMA Institute for Leprosy and Other Mycobacterial Diseases. The diagnosis was made by clinical criteria defined by World Health Organization (1998), according to which persons with one or more characteristic symptoms, such as hypopigmented or reddened skin lesion(s) with definite loss of sensation, and/or involvement of the peripheral nerves as demonstrated by palpable thickening with loss of sensation, and skin smear positivity for acid-fast bacilli, were included as leprosy patients. The skin smears from all leprosy patients were tested for the presence of acid-fast bacilli by Ziehl-Neelsen staining. Our study included 18 smear-positive MB patients and 38 smear-negative PB patients. All of these patients either were untreated or had started therapy less than one month earlier. Additionally, serum samples from 15 clinically active pulmonary tuberculosis patients, 14 other skin diseased patients and 20 healthy individuals were included as controls. Serum samples were obtained and stored at 20 °C until use. Approval for the study was obtained from our institute's ethics committee for scientific researches. Informed consent was obtained from the patients and controls.
Antibodies against PGL-I were detected by ELISA as described previously (Parkash et al., 2006). Our variations to this assay were that the serum samples added in the wells were incubated: for 2 h at 37 °C for the conventional assay, for 24 h at 4 °C for the one-step modified assay and for 2 h at 37 °C followed by 24 h at 4 °C for the two-step modified assay. To score a test as positive, cut-off points were determined using receiver operator characteristic curve analysis by calculating the percentage of true positives and false positives at various absorbance values obtained by ELISA. The absorbance value at which assay performance was optimal [(true positives+true negatives)/total sample number] was used as a cut-off point. The cut-off points for the conventional assay, one-step modified assay and two-step modified assay were 0.240, 0.400 and 0.375, respectively. A sample with an absorbance value of greater than the cut-off value in the respective assays was considered as positive. The differences between groups, with respect to the proportion of positively testing subjects, were assessed by chi-square test.
The number of positives and percentage positivity for all the assays are given in Table 1. On considering the results of all leprosy patients (MB+PB) together, the sensitivity of single-step modified approach was found to be higher than that of the conventional approach, but lower than that of two-step modified approach. However, the sensitivities of the three assays did not differ significantly (P>0.05) from one another. Surprisingly, one sample from a patient with a bacterial index (a criterion to estimate the bacterial density in the leprosy lesion) of 3+ was negative in all the assays. The failure to detect this patient could be due to immune-complex formation of anti-PGL-I antibody with the corresponding antigen, thus making the antibody unavailable for the assay. Specificity remained appreciably high (95.9100 %) and statistically similar (P>0.05) with all the assays. However, on further analysis, it became evident from the data that in case of two-step modified approach, the sensitivity with smear-positive MB patients was similar (94.4 %) to that of the conventional or one-step modified assay; whereas, in the smear-negative PB group it was (68.4 %) significantly higher (P<0.05) when compared to (36.8 %) the conventional assay but not significantly higher (68.4 vs 52.6 %; P>0.05) than the one-step modified assay. Added to this, all the samples positive by the conventional assay were also scored as positive by the one-step modified assay. Though the positivity with the one-step assay was higher than the conventional assay, the results of the two assays were statistically similar (P>0.05).