Comparative evaluation of staining techniques and polymerase chain reaction for diagnosis of intestinal microsporidiosis in immunocompromised patients.

CONTEXT: Microsporidia, which causes chronic diarrhoea in immunocompromised hosts, are often missed. The commonest diagnostic techniques include modified trichrome (MT) stain; however, it requires expertise and does not identify the species, which is important therapeutically. Other diagnostic techniques include Calcoflour white staining and polymerase chain reaction (PCR). Data on comparative utility of different diagnostic techniques are scanty. AIM: Comparison of Calcoflour white, MT staining and PCR for the diagnosis of intestinal microsporidiosis. SUBJECTS AND METHODS: Fecal samples of consecutive immunocompromised patients were evaluated for Microsporidia using Calcoflour white, MT stain and PCR. Species were identified by restriction fragment length polymorphism using HindIII and HinfI. Presence of Microsporidia by two or more techniques was considered true positive. Absence of Microsporidia by all three techniques was taken as true negative.
RESULTS: Of 730 patients, Microsporidia was detected in 28 (3.8%), 250 (34.2%) and 30 (4.1%) patients by MT, Calcoflour white stains and PCR, respectively. Enterocytozoon bieneusi was identified in all 30 (4.1%) patients. 30 (4.1%) and 479 (65.6%) patients were true positive and true negative, respectively. Sensitivity and specificity of Calcoflour white, MT stains and PCR were 100%, 93.8%, 96.8% and 68.5%, 100% and 99.8%, respectively. Diagnostic accuracy of MT stain and PCR was superior to Calcoflour white (99.6% vs. 69.8%; P < 0.05).
CONCLUSIONS: Though Calcoflour white stain is a highly sensitive, but it is nonspecific technique. MT stain and PCR with high sensitivity, specificity and diagnostic accuracy are useful diagnostic techniques. Furthermore, PCR is useful for species identification, which has therapeutic implications.

INTRODUCTION

Intestinal microsporidiosis causes chronic diarrhoea in immunocompromised hosts like human immunodeficiency virus (HIV) infected patients and renal transplant recipients worldwide.[12] Its common causative species are Enterocytozoon bieneusi (E. bieneusi) and Encephalitozoon intestinalis (Enc. intestinalis).[3] Encephalitozoon hellem (Enc. hellem) has also been reported in two cancer patients.[4] Its diagnosis is often missed as the spores of Microsporidia are variably small in size ranging from 1 to 4 μm. Earlier, transmission electron microscopy was considered gold standard for diagnosis of intestinal microsporidiosis; however, it is time consuming, costly and not available in most settings.[5] Presently, laboratory diagnosis depends upon direct visualization of spores in fecal samples by light microscopy using Weber's modified trichrome (MT) staining.[67] Fluorescent microscopy using Calcoflour white staining is a nonspecific diagnostic technique.[8] Both these staining techniques cannot differentiate between the species, which has important therapeutic implications as Enc. intestinalis respond well to currently used drug Albendazole whereas E. bieneusi respond variably to it.[9]

Polymerase chain reaction (PCR) based molecular techniques that amplify small subunit rRNA (SSU-rRNA) is being widely used for species specific detection of Microsporidia in immunocompromised hosts.[101112] This technique can detect as low as 102 spores/g of stool[13] and differentiate between species of Microsporidia.

Since there is scanty data on the comparative evaluation of MT, Calcoflour white stains and PCR for diagnosis of intestinal microsporidiosis in humans, we aimed to compare these techniques for diagnosis of intestinal microsporidiosis.

SUBJECTS AND METHODS

Study population

Consecutive patients with renal transplantation (RT), haematological malignancy (HM) and HIV infection, both with and without diarrhoea, referred to Parasitology division of department of Microbiology of a tertiary care centre for stool microscopy were evaluated for intestinal microsporidiosis during January 2006 to October 2013. HIV patients were recruited as per National AIDS Control Organisation guidelines, RT recipients with one or more transplantation and patients with any HM were included in the study. Patients with incomplete records were excluded from the study. The study protocol was approved by the Institutional Ethics Committee (PGI/DIR/RC/1085/2007).

Sample collection

Three consecutive stool samples were collected from each patient and subjected to microscopic examination. A part of stool was stored at − 40°C in normal saline for DNA extraction.

Sample processing

Stool samples of patients were subjected to formol ether concentration technique. The smears made from the pellet were air dried and fixed in methanol. For DNA extraction, samples of each patient were pooled in one aliquot.

Modified trichrome staining

Methanol fixed smears were stained using Weber's MT stain with some modifications.[6] Briefly, fast green was used as counter stain and the smear was incubated in trichrome stain at 50°C for 10–12 min. Each stained smear was observed under total magnification of ×1000.

Calcoflour white staining

Methanol-fixed smears were allowed to air dry, and a drop or two of the Calcoflour solution was added. After 2 or 3 min at room temperature, the slides were rinsed with slow-running water and counter stained with 0.1% Evan's blue (Sigma; E-2129) in Tris-buffered saline (pH 7.2) for 1 min at room temperature.[14] The slides were rinsed under slow-running water, allowed to air dry, and viewed under a ultraviolet microscope at a wavelength of 395–415 nm (observation light of 455 nm). Organisms appeared as bluish-white or turquoise oval halos.

Polymerase chain reaction amplification

DNA was extracted from pooled stool samples using QIAamp Qiagen mini stool kit (Qiagen Inc., Valencia, CA, USA) according to manufacturer's instruction with some modifications. Briefly, the sample was suspended in phosphate-buffered saline and subjected to centrifugation. The suspension was heated at 80°C for 10 min. The extracted fecal DNA was subjected to amplification of the conserved region of SSU rRNA gene of E. bieneusi and Encephalitozoon species using previously published forward primer C1 (5’CACCAGGTTGATTCTGCC-3’) and reverse primer C2 (5’GTGACGGGCGGTGTGTAC-3’) as per published protocol.[15] Amplified products were electrophoretically analyzed on agarose gel and stained with ethidium bromide.

Species identification

The amplified fragments were subjected to digestion using restriction endonucleases HinfI and HindIII to differentiate between four human Microsporidia species, E. bieneusi, E. intestinalis, E. cuniculi and E. hellem.[15] 10 μL of amplified DNA was digested with 5 U of HinfI and HindIII in a final volume of 15 μL. Digested fragments were analyzed on 2.5% agarose gel stained with ethidium bromide. Amplified fragments were analyzed electrophoretically on agarose gel.

Statistical analysis

For comparative evaluation of three techniques, presence of Microsporidia by two or more techniques was considered as true positive. Absence of Microsporidia by all three techniques was considered true negative. Sensitivity, specificity, positive predictive value, negative predictive value and diagnostic accuracy were calculated as per standard formulae.

RESULTS

Result of microscopy

Of 730 patients included in the study, 195 (26.7%) had HIV, 148 (20.3%) had HM and 387 (53%) underwent RT. Of them, 28 (3.8%) and 250 (34.2%) patients were positive for Microsporidia by MT and Calcoflour white stains, respectively. Figure 1a and b shows the representative picture of spores of Microsporidia using MT and Calcoflour white staining, respectively.

Figure 1

Representative picture of spores of using (a) modified trichrome stain under × 1000, (b) Calcoflour white stain

Polymerase chain reaction-restriction fragment length polymorphism for species identification

Polymerase chain reaction was positive for Microsporidia in 30/730 (4.1%) patients and restriction fragment length polymorphism (RFLP) using restriction enzymes Hinf I and Hind III identified E. bieneusi as the causative species of intestinal microsporidiosis in all of them. E. intestinalis, E. cuniculi and E. hellem were not identified in any patient. All the positive samples were confirmed by sequencing. Figures 2 and 3 show the representative gel pictures of patients positive for Microsporidia and enzymatic digestion of amplified products using restriction enzyme Hinf I and Hind III.

Figure 2

An agarose gel showing polymerase chain reaction products of human fecal samples positive for . Lane M, 50 bp ladder; Lane PC, positive control for Lane 1-3, positive for Microsporidia; NC, Negative control

Figure 3

Digestion pattern of polymerase chain reaction products using restriction enzyme HinfI and Hind III. Lane M, 100 bp ladder; Lane PC, positive control for using HinfI; Lane 2 and 4, samples positive for E. bieneusi using HindIII; Lane 3, samples positive for E. bieneusi using HinfI

Sensitivity, specificity, positive and negative predictive values and diagnostic accuracy of different techniques

For comparative evaluation of these methods, 30 (4.1%) and 479 (65.6%) patients were considered true positive and true negative, respectively. Samples positive for Microsporidia by three techniques are shown in Table 1. Sensitivity of Calcoflour white stain, PCR and MT stain was 100%, 96.8% and 93.8%, respectively [Table 2]. Specificity of MT stain was 100%, PCR: 99.8% and Calcoflour white stain: 68.5% [Table 2]. The specificity of MT stain and PCR was higher than that of Calcoflour white stain (P < 0.05). Similarly, diagnostic accuracy of MT stain and PCR was superior to that of Calcoflour white stain (99.6% vs. 69.8%; P < 0.05).

Table 1

Detection of microsporidia by three techniques (n=730)

Table 2

Comparative evaluation of calcoflour white, MT stains and PCR for the detection of microsporidia

DISCUSSION

In the present study, Calcoflour white, MT stains and PCR showed high sensitivity to diagnose intestinal microsporidiosis in immunocompromised patients. Calcoflour white stain had poor specificity and diagnostic accuracy than that of MT stain and PCR. Furthermore, PCR-RFLP identified E. bieneusi in all patients infected with Microsporidia.

In the present study, Calcoflour white stain showed 100% sensitivity, which is in accordance with previously published data.[16] Conversely, in a recent study, sensitivity of Calcoflour white stain was reported to be 79.7%.[5] Specificity of Calcoflour white was 68.5% in the present study, which is close to the specificity reported in other studies ranging from 77.4% to 82.2%.[58] Calcoflour white stain binds to the chitin of the endospore layer of Microsporidia, which is also present on other fungi, therefore it gives false positive results. This could be a probable cause of high sensitivity but poor specificity. Therefore, fluorescent microscopy using Calcoflour white stain is not a suitable method for detecting spores of Microsporidia in fecal sample.

Microsporidia stained with MT stain, appears bright pink against green background. Often a central diagonal stripe is visible within the spore with an empty posterior vacuole.[8] In the present study, MT stain showed a sensitivity of 93.8%. Many previously published studies have reported variable sensitivity of MT stain ranging from 54% to 100%.[58131617] This technique is possible to perform in a small setup but its sensitivity principally depends upon the technical expertise as the spore size is very small and could be easily missed by the microscopist. Also, the initial standardization of staining takes time due to thick spore wall. The specificity of this staining technique was 100% in the current study which was comparable to specificities reported earlier ranging from 82.8% to 100%.[58131617] The diagnostic accuracy of MT stain is also higher as compared to Calcoflour white stain. The spores of Microsporidia are easily demarcated from yeast cells in MT stain.[8] Therefore, it is easy to differentiate between the two, which is difficult in case of Calcoflour white stain. Therefore, light microscopy using MT stain could be used as a diagnostic tool in laboratories with small setup.

Though light microscopy is an economical method to be performed in a small setup, but it cannot differentiate between different spp. of Microsporidia, which is important for therapeutic purposes, epidemiological and genotypic studies. In view of this, molecular methods involving PCR is a powerful tool to differentiate between species in a clinical sample. In the present study, sensitivity of PCR was 96.8%, which is in accordance with the earlier studies reporting it to be up to 100%.[517] In another study on blinded, externally controlled multicentre evaluation of PCR for detection of E. bieneusi in stool samples, authors reported sensitivity ranging from 71% to 100%.[13] In the present study, PCR could not pick up one case diagnosed as intestinal microsporidioisis by MT stain. This could be related to the presence of PCR inhibitors in the stool samples like bile salts, complex polysaccharides etc., However, PCR is a robust method to detect as low as 102 spores/g of sample.[13] Furthermore, in the present study, PCR-RFLP identified E. bieneusi as the commonest species infecting all patients. Other species, namely, E. intestinalis, E. cuniculi and E. hellem were not detected in the current study. However, in an earlier study from India, Enc. intestinalis was the commonest species identified followed by E. bieneusi.[18] Therefore, PCR is a good diagnostic technique, which can differentiate between the species and thus, has therapeutic implications.

CONCLUSION

Calcoflour white stain is a highly sensitive but nonspecific technique for the detection of Microsporidia. MT stain and PCR with high sensitivity, specificity and diagnostic accuracy are useful diagnostic techniques. Furthermore, PCR is useful for species identification, which has therapeutic implications.