Introduction

Cryptosporidium spp. and Giardia spp. are parasitic protozoa responsible for gastrointestinal illness in several animal species as well as in humans [1, 21], which includes opening the cartridge and cutting out the membrane. The membrane pieces were resuspended in 200 µL of PBS 1X buffer and vortexed one minute to allow the cells to be collected in the liquid.

Detection and enumeration of (oo)cysts

Concentrates either from the IMS or the microfiltration were then fixed on a microscope slide and stained with either EasyStain kit (BioPoint Inc., Australia) or Aqua-Glo™ G/C kit (Waterborne Inc., United States) following the manufacturer's instructions as required by the USEPA Method 1623.1. The slides were analyzed by fluorescence microscopy with a Zeiss Axio Observer Z1 microscope connected with an Axiocam MRm camera (Carl Zeiss, North York, ON, Canada) to enumerate the number of (oo)cysts collected. The techniques were compared by calculating the percentage of (oo)cysts collected by each one.

Statistical analysis

Each combination of steps was done in biological triplicate. Statistical analyses were made with the R package RCommandr version 3.5.0. The normality of the recovery values for each condition was determined by a Shapiro–Wilk test. Averages obtained for each condition and each parasite were then compared by the Kruskal–Wallis non-parametric test with a threshold of 5%. The data for the conditions applying to a normal distribution were compared together with an ANOVA test with a threshold of 5%.

Results

Figure 2 presents the results obtained with the various combinations of techniques. Although none of these alternative approaches gave significantly superior recovery rates to USEPA Method 1623.1, equivalent results were obtained in the case of Cryptosporidium results. It was determined that the medians of the data generated with the four approaches were not statistically different for this parasite. For Giardia, the USEPA Method 1623.1 produced recovery rates with a median statistically significantly higher than the three other approaches tested. Even if the distribution of data could look otherwise, averages of the data obtained for Cryptosporidium and Giardia with the USEPA Method 1623.1 were not statistically different. But this global behavior of higher recoveries for Giardia than for Cryptosporidium has been seen in previous studies as well [17, 22, 23]. Only the comparison of data from the USEPA protocol for Giardia and the blender elution followed by IMS gave statistically significantly different averages, with the USEPA method giving higher values.

Fig. 2
figure 2

Recovery of Cryptosporidium and Giardia according to the combination of techniques applied. Condition 1 consists of the complete protocol according to USEPA Method 1623.1. Condition 2 is the combination of the elution according to the USEPA with the concentration by microfiltration. Condition 3 is the alternative elution protocol combined with the concentration by IMS. Condition 4 is the alternative elution protocol paired with the concentration by microfiltration. The symbol ■ states that the averages are statistically significantly different with a threshold of 5%. The symbol □ indicates that the averages are non-statistically significantly different with a threshold of 5%. The symbol ○ means that the medians are non-statistically significantly different with a threshold of 5%

Full size image

The costs and the laboratory work time required for each approach are summarized in Table 1 and detailed. Some of the alternative approaches allow a gain of time of a few hours of laboratory work. The costs can be significantly reduced from 100 to 650 CAD depending on the method chosen compared to the USEPA Method 1623.1. It could be advantageous especially in the case of Cryptosporidium where all techniques generated non-statistically significant different results. Despite the decrease in costs, the alternative elution paired with the microfiltration does not seem to be the best approach to adopt, given the low recoveries obtained. However, USEPA elution with microfiltration appears to be the most advantageous combination of alternative approaches regarding the time gain and the decrease of costs for a similar recovery. This is particularly promising in a research context.

Table 1 Costs and time required per sample for each combination of techniques tested in this study
Full size table

Limitations

Some limitations in the approaches tested in this study are worth mentioning. First, our inexperience with the USEPA Method 1623.1 protocol likely may have contributed to the relatively low recovery rates. However, low recovery rates were also reported in the past by interlaboratory validation assays led by the USEPA [11], and our recovery rates align with those. With experience and practice, higher recoveries might be expected. Typically, only well-trained personnel adept at obtaining higher recoveries are allowed to analyse samples from clients in environmental analysis laboratories, while the present study was done in an academic research context.

Second, with the alternative elution protocol tested during this study, it was not possible to physically retrieve the entire filtration membrane from the EnviroChek HV cartridge following its opening with a pipe cutter. Approximately 30% of the filtration surface remained covered by the polycarbonate housing despite all our efforts to cut the cartridge closer to the extremity (Additional file 1, Figure S1). Therefore, the parasites stuck to that portion of the filtration surface could not be collected with this approach.

Third, although the concentration of parasites by IMS is expected to remove all other cells that do not belong to either Cryptosporidium or Giardia genera, our experiments showed that does not always occur. Many bacteria cells could also be seen (Additional file 1, Figure S2). Our experiments were done by artificially spiking 10 L of water with 500 000 bacterial cells to mimic contaminated environmental water. However, some raw water samples may contain even higher concentrations of bacteria. For example, the River Ruhr in Germany, which was studied by Strathmann et al. [24], contained about 3.4 × 106 total cells per mL (more than 50 000 times more bacteria than in our own samples). We conclude that the problem of interfering bacteria carried over during the IMS could be quite cumbersome for the analysis of some environmental samples and could interfere with the analysis. IMS beads can also confuse Cryptosporidium oocyst detection if they are carried over in the sample until the microscopic examination (see Additional file 1, figure S3 for comparison).

Finally, centrifugation is a major issue to consider in this protocol to improve parasite recovery. The USEPA Method 1623.1’s centrifugation step is at 1500 × g for 15 min. In a previous study, this centrifugation caused a loss of 8 to 14% of cells when compared with the same sample composition submitted only to IMS and fluorescence microscopy [25]. Lechevallier et al. [26] found that higher centrifugation speed helped to recover more (oo)cysts; therefore, to increase the proportion of (oo)cysts collected, we chose a centrifugation speed of 5855 × g for 30 min following our alternative elution technique. Since Lechevallier et al. (26) demonstrated that even a speed of 17 300 × g would not disrupt them, it is unlikely that the centrifugation applied here would have broken them. We did not evaluate the loss of parasites at each step in the present study.