CASE NOTES
Coxiella burnetii – A seminal case study
Shayne Fell, Ronald Coilparampil, Yan Huang, Cheryl Jenkins, Keith Walker, Kristie Harper and Ian Marsh
Elizabeth Macarthur Agricultural Institute, NSW Department of Primary Industries, Menangle, NSW
Posted Flock & Herd September 2021
INTRODUCTION
Coxiella burnetii is a zoonotic pathogen that occurs worldwide, except for New Zealand, causing Query (Q) fever in humans and coxiellosis in animals. In cattle, sheep and goats, C. burnetii is most commonly associated with sporadic late-term abortions and dead or weak young. In cattle it has also been associated with infertility. In animals the infection is often persistent, and the bacteria can be shed in a variety of secretions and excreta.
HISTORY
In July 2019, the diagnostic laboratory at EMAI was contacted by a submitter from an artificial insemination (AI) centre with concerns about results from recent coxiellosis testing. Their herd had no previous history of coxiellosis; however, two animals had returned positive results by CFT and ELISA. Samples from the same animals had also been sent to a second laboratory for testing but yielded negative results. This troubled the client and based on no previous knowledge of coxiellosis on the property and negative results from the other laboratory, led them to believe there may have been an issue with the results obtained from EMAI. Furthermore, no animals had been introduced on to the property as they operated as a "closed herd". However, the client was keen to determine if there really was a coxiellosis problem within their herd. The willingness of the client to co-operate was important in the event of the final outcome from our investigation. Further testing on subsequent animals from the same property yielded two additional animals to be positive for coxiellosis. Three of the positive animals, in total, were bulls and the fourth was a teaser steer used during the collection process. This paper discusses the systematic approach taken to investigate this case and how a cooperative investigation resulted in a positive outcome for the client and EMAI laboratory.
INITIAL TESTING AND INVESTIGATION AT EMAI
To confirm the initial results for the samples submitted to EMAI, repeat testing was undertaken by CFT and ELISA. The results, in most cases, were consistent (Table 1). Only one sample, Bull AI (21/6/19), produced a notably different result when the initial and repeat results were compared, that being suspect and positive, respectively.
Table 1: Initial investigation of serology test results of 2 runs done on different days. (ELISA cut-off points Negative <30, Suspect between 30 and 40, Positive >40: CFT <8 negative, 8 or higher positive) (NT=not tested)
| Animal | Date of Collection | QF CFT results | QF ELISA results | ||
|---|---|---|---|---|---|
| Initial | Repeat | Initial | Repeat | ||
| Steer AR | 29/04/2019 | NT | 16 | 60 | 64 |
| Steer AR | 28/05/2019 | NT | 32 | 280 | 299 |
| Steer AR | 21/06/2019 | <8 | 8 | 218 | 246 |
| Bull MR | 17/04/2019 | <8 | <8 | 55 | 83 |
| Bull MR | 16/05/2019 | <8 | <8 | 122 | 140 |
| Bull AI | 21/06/2019 | <8 | <8 | 32 | 60 |
COMPARISON BETWEEN EMAI AND SECOND LABORATORY RESULTS
To understand the difference between the EMAI results and the second laboratory results (all negative; data not shown), the second laboratory was contacted to find out what differences may exist between the methods they were using for both CFT and ELISA testing, these are summarised in Table 2.
Table 2: Comparison of methods used between Laboratories.
| Laboratory | CFT | ELISA |
|---|---|---|
| EMAI | Cold Fixation | IDEXX QF ELISA |
| Second Laboratory | Warm Fixation | ID Vet Multi species QF ELISA |
With respect to ELISA, previous investigations by the EMAI laboratory have shown the ID Vet QF ELISA is less sensitive than the IDEXX QF ELISA but specificity is comparable for both ELISA (data not shown).
To better understand the impact of the fixation methods on the CFT, we compared the two methods and concluded that the cold method is more sensitive (Table 3). It could detect positive sample out to a 1/256 working dilution, whereas, the warm method could detect out to a 1/32 working dilution. This makes the cold method is three titres more sensitive than warm method.
Given that the same antigen (supplied by r-biopharm) is being used at EMAI and at the second laboratory for the CFT tests, it can be concluded that the different fixation methods is most likely the reason for the disagreement in the results produced between from the two laboratories for this clients samples.
These results were communicated to the client outlining the EMAI laboratories reasoning behind the different results between the two laboratories. To help progress the investigation the client forwarded EMAI additional information about the previous testing history of all four animals in question.
Table 3: Sensitivity comparison between Cold and Warm CFT methods, testing carried out in duplicate.
| QF CFT Sample Working Dilution |
Warm Method Titre (duplicate) |
Cold Method Titre (duplicate) |
||
|---|---|---|---|---|
| 1 | 2 | 1 | 2 | |
| Neat | 256 | 256 | 1024 | 1024 |
| 1/2 | 64 | 64 | 512 | 512 |
| 1/4 | 32 | 32 | 256 | 256 |
| 1/8 | 16 | 32 | 128 | 128 |
| 1/16 | 16 | 16 | 64 | 64 |
| 1/32 | 8 | 8 | 32 | 32 |
| 1/64 | <8 | <8 | 16 | 16 |
| 1/128 | <8 | <8 | 16 | 8 |
| 1/256 | <8 | <8 | 8 | <8 |
| 1/512 | <8 | <8 | <8 | <8 |
| 1/1024 | <8 | <8 | <8 | <8 |
| 1/2048 | <8 | <8 | <8 | <8 |
CONTINUED INVESTIGATION WITH NEW INFORMATION
Analysing this testing history, it was interesting to note that all of the positives have occurred from mid-April to mid-June 2019 which appeared to be indicative of spread within the herd.
To confirm the status of all samples, both ELISA methods were compared (Table 4). Whilst positive results were achieved with both tests, the IDVET ELISA once again appeared to be less sensitive than the IDEXX ELISA. Furthermore, the IDEXX results even indicated a possible progression of disease over the time course of samples for the steer AR and bull MR, these being the two animals initially detected as positive by the EMAI laboratory.
Table 4: Summary of results comparing IDEXX and IDVET ELISA and showing change in results over time. (AR-steer and MR-bull were initial 2 animals detected as positive by serology) (NT=not tested)
| Bull ID | Collection Date | IDEXX ELISA | Result | IDVET ELISA | Result |
| AR (steer) | 29/04/2019 | 60 SP% | Positive | 15 SP% | Negative |
| AR (steer) | 25/05/2019 | 280 SP% | Positive | 66 SP% | Positive |
| AR (steer) | 21/06/2019 | 218 SP% | Positive | 51 SP% | Positive |
| MR (bull) | 17/04/2019 | 55 SP% | Positive | 17 SP% | Negative |
| MR (bull) | 29/04/2019 | 46 SP% | Positive | 15 SP% | Negative |
| MR (bull) | 16/05/2019 | 122 SP% | Positive | 40 SP% | Suspect |
| VI (bull) | 17/4/2019 | 15 SP% | Negative | NT | |
| VI (bull) | 5/6/2019 | 42 SP% | Positive | 9 SP% | Negative |
| VI (bull) | 21/06/2019 | 29 SP% | Negative | 11 SP% | Negative |
| AI (bull) | 17/04/2019 | 6 SP% | Negative | NT | |
| AI (bull) | 5/06/2019 | 36 SP% | Suspect | 8 SP% | Negative |
| AI (bull) | 21/06/2019 | 32 SP% | Suspect | 10 SP% | Negative |
POLYMERASE CHAIN REACTION
As part of the investigation undertaken at the second laboratory, PCR was performed on serum samples at the client's request and in all cases the results were DNA consistent with C. burnetii, not detected. These PCR results further confused the client and gave reason to question the validity of EMAI QF CFT and ELISA results. However, serum is most likely not the best sample for PCR and if PCR were to be used it would be better to assess semen samples directly.
Following communication with the client and Q Fever experts at the Australian Rickettsial Reference Laboratory (ARRL), we recommended the client send semen samples to the ARRL for PCR. The client agreed to submit semen samples collected from two of the four animals in question around the time when the serum antibodies were detected by serological tests. Only two bulls could be tested, as one of the bulls had no semen available from that period and the last animal was a steer. Semen was sent directly to the ARRL for PCR testing and both animals tested were found to be positive for DNA consistent with C. burnetii (Table 5).
Table 5: PCR Results for the 2 bulls tested, animals are considered positive if DNA is detected in at least one of the PCRs
| Bull ID | Collection Date | com1 | hptAB | IS1111 | Overall Result |
| MR | 12/3/19 | Undetected | Positive | Positive | Positive |
| MR | 17/4/19 | Undetected | Positive | Positive | Positive |
| MR | 29/4/19 | Undetected | Undetected | Positive | Positive |
| MR | 6/5/19 | Undetected | Undetected | Positive | Positive |
| MR | 16/5/19 | Undetected | Positive | Positive | Positive |
| MR | 6/6/19 | Undetected | Positive | Positive | Positive |
| VI | 14/8/19 | Undetected | Positive | Positive | Positive |
DISCUSSION
From this investigation it became apparent that the original results supplied from EMAI to the client were in fact correct. The disparity between the two laboratories can be explained by (1) different fixation methods for CFT, (2) different ELISA tests and (3) the sample used for PCR. Interestingly, when PCR was used on an appropriate sample type, we were able to provide the client with further evidence of the presence of C. burnetii within their herd(s). This highlights the importance of understanding the correct sample type as a misleading result can often be obtained by testing of an inappropriate sample.
The client initially doubted the results on the basis of conflicting results, no previous evidence of infection but equally as they operated a closed herd. However, in the case of C. burnetii there are a number of other possible sources of infection and factors outside of herd management may need to be considered.
It became apparent from discussions during the investigation that not all animals were from the same property and in fact were located on three separate properties geographically distinct but close to each other. Therefore, direct transmission between animals was not the likely cause of these animals being exposed to C. burnetii. It would appear there was an environmental source of infection.
Other potential sources of transmission were discussed, and the client informed us that there was a large population of rabbits in that area around the time, and there had been a cull carried out around the period that infection was first detected. It has been reported that the European rabbits being a reservoir for C. burnetii (Gonzalez-Barrio et al. 2015). High individual and population seroprevalences were observed in wild and farmed rabbits. Many species of wildlife and domestic animals (CDNA 2018) have been shown to harbour C. burnetii and the rabbit situation suggested here is just one of many possible sources of infection into an otherwise closed herd.
Furthermore, infection with C. burnetii can occur by breathing in the bacteria that is in the air or dust. Infection may occur after direct exposure to infected animals and their products (placenta, abortion products, hides, wool, manure, straw etc.), especially at the time of parturition or slaughtering. Because C. burnetii may persist for prolonged periods in the soil, these aerosols may also be produced long after the release of bacteria by infected animals. Moreover, bacterial aerosols can be spread for long distances by the wind, resulting in C. burnetii cases far away from the primary contaminated areas (Tissot-Dupont et al. 1999). Therefore, while the notion of a closed herd may be relevant in terms of transmission of other animal to animal infectious agents, with respect to C. burnetii this is not always the case, give that it is ubiquitous.
ACKNOWLEDGEMENTS
The authors would like to thank the Australian Rickettsial Reference Laboratory for advice with respect to the PCR testing protocols.
REFERENCE LIST
- Gonzalez-Barrio D, Maio E, Viera-Pinto M, and Ruiz-Fons F. European Rabbits as Reservoir for Coxiella burnetii. Emerging Infectious Diseases 2015;21:1055-1058
- Q-Fever: CDNA National Guidelines for Public Health Units (2018) Communicable Diseases Network Australia
- Tozer SJ, Lambert SB, Strong CL, Field HE, Sloots TP, Nissen MD. Potential animal and environmental sources of Q fever infection for humans in Queensland. Zoonoses Public Health 2014;61(2):105-12
- Tissot-Dupont H, Torres S, Nezri M, and Raoult D. Hyperendemic Focus of Q Fever Related to Sheep and Wind. American Journal of Epidemiology 1999;150(1):67-74