Table C.1.
Detection methods for Cryptosporidium in food
| Type of food | Methods and their validation | |||||||
|---|---|---|---|---|---|---|---|---|
| Standard method | Other suitable methods | |||||||
| Fresh produce and herbs |
ISO 18744, 2016 Microbiology of the food chain — Detection and enumeration of Cryptosporidium and Giardia in fresh leafy green vegetables and berry fruits. Based on elution from the surface of the leaf or berry, concentration of oocysts from the eluate by immunomagnetic separation (IMS), and enumeration by immune fluorescence microscopy (IFM) LOD: no data Validated in a ring trial (8 data points/labs): |
Alternative of ISO according Utaaker et al. (2015) Method is using a smaller volume of magnetic beads in the IMS step and buffers made in‐house LOD: no data Validated in a ring trial (7 data points/labs): |
Oocyst elution and concentration by ISO or Utaaker et al. (2015) and detection by PCR (Hohweyer et al., 2016) Using PCR for detection LOD: 3 oocysts/g basil (30 g seeded with xx oocysts) < 1 oocyst/g raspberries Validation in one laboratory study |
|||||
| Lettuce | Raspberries | Leafy greens | Basil | Raspberries | ||||
| Mean recovery rate (100 oocysts) | 30.4% | 44.3% | Mean Recovery rate (50 oocysts) | 53.0% | Mean recovery rate (408 oocysts) | 11.0% (6–23%) | 14% (1–45%) | |
| Sensitivity | 89.6% | 95.8% | Sensitivity | 87.5% | ||||
| Specificity | 85.4% | 83.3% | Specificity | 87.5% | ||||
| Accordance | 82.4% | 92.1% | Accordance | n/a | ||||
| Concordance | 81% | 91.8% | Concordance | 80.0% | ||||
|
Advantages: Quantitative Microscopy slides amenable to onward testing by PCR for species and genotype Disadvantages: Expensive Time consuming Species non‐infective for humans will be counted Further processing is needed for molecular characterisation Recovery of oocysts from sample matrix can be low No viability or infectivity assessment is possible |
Advantages: Significantly cheaper method than ISO Disadvantages: As for ISO, apart from cost |
Performance will depend on DNA extraction and PCR efficiency and specificity Advantages: High throughput detection Potential for species identification Disadvantages: Not quantitative Sample preparation remains time consuming |
||||||
| Fruit and Vegetable juice | No |
IMS‐IFM Orange juice LOD 5–50 oocysts/10 mL (Frazar and Orlandi, 2007) IMS‐IFM Apple juice LOD 5–50/10 mL (Frazar and Orlandi, 2007) IMS‐IFM Apple juice LOD 10/100 mL (Deng and Cliver, 2000) IMS‐PCR LOD Apple juice 30/100 mL (Deng and Cliver, 2000) No ring trials |
Microfiltration‐PCR LOD 10/250 mL apple juice (Minarovičová et al., 2010) | |||||
| Dairy: Milk | No |
IMS‐PCR LOD Homogenised milk 10/100 mL (Deng and Cliver, 2000) IMS‐PCR Raw and pasteurised whole milk 10/50 mL (Di Pinto et al., 2002) IMS‐nPCR whole milk 5–50/10 mL (Frazar and Orlandi, 2007) Centrifugation‐PCR raw milk 1–10/20 mL (Laberge et al., 1996) |
Microfiltration‐single‐tube nested real‐time PCR LOD 10/100 mL milk (Minarovičová et al., 2011) | |||||
| Dairy: Fermented products | No | No data | ||||||
| Cheese | No | No data | ||||||
| Molluscan shellfish | No |
Sieved, pooled tissue homogenates (a method most commonly used), processed by IMS and detection by IFM or PCR LOD: no data Validation: no ring trials Data from systematic study (MacRae et al., 2005): known numbers of oocysts seeded into a 20‐L tank of seawater and circulated for 20 min to disperse before either 10 mussels, 2 oysters or 2 scallops were placed in the tank for 4 h at 5°C Shellfish removed and tissue homogenates (30 s in a Waring blender) were pooled for each sample and processed (5 tests per sample, aggregate 0.5‐mL tissue) by centrifugation, IMS and IFM Recovery rates were: Mussels: 34% of 20,000, 12% of 2,000, 20% of 200 oocysts spiked into the tank Oyster 69.5% of 20,000, 60.5% of 2,000, 48% of 200 oocysts spiked into the tank Scallop 32.5% of 20,000, 30% of 2,000, 65% of 200 oocysts spiked into the tank Gómez‐Couso et al. (2006) compared IFM and PCR detection, using 10% of spiked mollusc sediments containing 0, 10, 50, 100, 500 and 1,000 C. parvum oocysts. Average numbers of oocysts detected in three replicates by IFM were 0, 0.7, 3.6, 6.7, 37.6 and 77.2, respectively; the average percentage of recovery was approximately 70.0% and did not differ by spiking dose DNA was extracted and 10 amplification reactions were performed for each spiking dose. The theoretical numbers of oocysts present in the volume used in the PCR technique were 0, 1, 2, 4 and 40 oocysts. PCR positive results diminished as spiking dose decreased, ranging from 90% for 40 oocysts to 10% for 1 oocyst |
Pepsin digestion of 3 g pooled homogenate using IFM for detection (Robertson and Gjerde, 2007) LOD: no data Validation: no ring trials Data from developing laboratory: 68–79% of 179 oocysts (horse mussel and oyster homogenates) This method has been used by the developing laboratory and there is one published report of a sample survey that has used it (Aguirre et al., 2016) |
|||||
| Meat | No |
Surface elution centrifugation, IMS, IFM (Robertson and Huang, 2012) LOD, not stated Validation: No ring trials Recovery rates for ~ 100 oocysts were 63.5% (CI 54.6–70.2) |
||||||
ISO: International Organization for Standardization; LOD: limit of detection; PCR: polymerase chain reaction.
Sensitivity (=true positive rate): the ability of a test to correctly identify positive samples (spiked or contaminated samples).
Specificity (=true negative rate): the ability of a test to correctly identify negative samples (non‐inoculated samples)
Accordance or repeatability: Accordance is the percentage (ratio) that two identical test materials analyzed by the same laboratory under standard repeatability conditions will both be given the same result (i.e. both found positive or both found negative) (Langton et al., 2002).
Concordance or reproducibility: Concordance is the percentage (ratio) that two identical test materials sent to different laboratories will both be given the same result (i.e. both found positive or both found negative result) (Langton et al., 2002).