Background: Giardiasis is an intestinal diarrhoeal illness caused by the flagellate protozoan parasite Giardia intestinalis (synonymous with Giardia lamblia and Giardia duodenalis). Traditionally, giardiasis has been diagnosed in patients using faecal concentration and microscopy techniques. Non-microscopy based tests available for the laboratory diagnoses of giardiasis include recent innovations in polymerase chain reaction (PCR) and immunoassays with increased sensitivity. The laboratory diagnosis of giardiasis is complicated by the intermittent excretion of the parasite and asymptomatic presentation that sometimes occurs with this infection. Clinicians may on occasion treat patients for giardiasis on clinical suspicion alone when diagnostic tests have failed to identify Giardia intestinalis and some of the patients do get better putting into question the performance of the diagnostic test used. At the Hospital for Tropical Diseases (HTD) in London the ova, cysts and parasite microscopy (OCP-M) is the front line test for diagnosing giardiasis.
Aim: The aim of this study was to critically analyse the performance of a commercial and a published real-time PCR diagnostic tests for their potential use as front line tests for the diagnosis of giardiasis in the clinical parasitology diagnostic laboratory at the HTD. Storage conditions that will allow the best yield of Giardia intestinalis DNA from stored faecal samples were also investigated in this study.
Methods: In the absence of a gold standard, a composite reference standard (CRS) of enzyme immunoassay (EIA) and rapid membrane test (RMT) was used to evaluate the performance of Primerdesign Ltd. real-time PCR kit for Giardia intestinalis (which detects only assemblages A and B subtypes) and a real-time PCR assay using Verweij et al published primers (Verweij real-time PCR) which targeted the (SSU) rRNA gene. The two tests were compared with the OCP-M test in a diagnostic accuracy study using a non-probability sampling technique with consecutive samples.
Results: The Verweij real-time PCR which targeted the (SSU) rRNA gene showed a diagnostic sensitivity of 93.4 % (95 % CI: 86.2 to 97.5 %) and a specificity of 74.7 % (95 % CI: 63.6 to 83.8 %) with a limit of detection (LOD) of < 5 cysts/ml. The Primerdesign Ltd. real-time PCR which also targeted the gdh gene showed a diagnostic sensitivity of 61.5 % (95 % CI: 50.8 to 71.6 %) and specificity of 98.7 % (95 % CI: 93.2 to 99.8 %) with a limit of detection (LOD) of ≤ 114 cysts/ml. Also, with a serially diluted 1 in 10 dilutions of a known concentration Giardia intestinalis DNA solution, the Verweij real-time PCR produced efficiency (E) of 96 % (the slope was -3.414) with a correlation coefficient (R2) of 0.99 and a copy number variance predominantly less than 10 % (< 10 %). The Primerdesign Ltd. had E = 100 % (the slope was -3.342), R2 = 0.95, and a copy number variance predominantly greater than 10 % (> 10 %).
In this study, the OCP-M missed 16.5 % Giardia positive stool samples contrasted with 6.6 % missed by the Verweij real-time PCR. The Verweij real-time PCR therefore showed approximately 10 % increase (i.e. 16.5 % - 6.6 %) in detection rate over the OCP-M and with an estimated detection limit of < 5 cysts/ml of stool, it also correctly identified 70 % (14/20) of the discrepant cases as true positives. OCP-M identified 10 % (2/20). When sensitivities were adjusted for the Verweij real-time PCR as a result of enhancement in the detection rate of the CRS, 19.3 % (94.3 % - 75 % = 19.3 %) more positive cases were noted. The Verweij real-time PCR proved to be more robust than the OCP-M and the Primerdesign Ltd. PCR and has therefore been shown to be more suited for deployment as a first line diagnostic test than the other two index tests. Even in combination with the OCP-M, the sensitivity remained unchanged at 93.4 %. With its high specificity, the Primerdesign Ltd. Giardia PCR kit may be useful for partitioning clinical history for epidemiological studies but with LOD of ≤ 114 cysts/ml of stool and R2 < 0.99 when faecal samples are involved, it will require further optimisation for use on clinical samples. Up to the end of April 2013, a literature search showed no independent evaluation of this Giardia real-time PCR kit.
Storage affects molecular analyses and from the findings of this study, stool samples are best stored in industrial methylated spirit and kept at 4-6 °C if they are to be used for real-time PCR for Giardia intestinalis detection. Alternatively they can be stored in the freezer at -20 °C without industrial methylated spirit. Samples should however be tested within three months of storage.
Conclusion: The reason why some patients get better when they are treated empirically following microscopy negative results for Giardia intestinalis may be found in the fact that, in this study, the OCP-M failed to detect 16.5 % of positive cases. The Verweij real-time PCR performed better than the OCP-M and showed an improvement of 10 % in Giardia intestinalis detection rate. The Primerdesign Ltd. Giardia PCR kit requires further optimisation for use on clinical samples. The Verweij real-time PCR was more robust than the OCP-M and the Primerdesign Ltd. PCR and therefore is more suited for use as a first line diagnostic test with best results obtained when stool samples are first treated with industrial methylated spirit, stored in the fridge at 4-6 °C and tested within three months of storage.
The Verweij real-time PCR assay may be used as a standalone test for in combination with the OCP-M, there was no improvement in the 93.4 % sensitivity when it was used alone. The OCP-M, however, has the advantage of identifying the presence of other parasites.
|Date of Award||Jun 2013|
|Supervisor||Sally Anne Kilburn (Supervisor) & Graham Mills (Supervisor)|