SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials And Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Objectives: To describe a method of faecal smear production and to determine the sensitivity and specificity of faecal smear for detection of parasitic larvae using Baermann sedimentation as a gold standard

Methods: Faecal smears were produced from samples submitted to the Royal Veterinary College Diagnostic Laboratory Service for Baermann sedimentation. An inexperienced and an experienced assessor each examined the smear for larvae for a maximum of five minutes.

Results: One hundred and eighty six samples were analysed of which 28 were positive for Angiostrongylus vasorum on Baermann sedimentation. The experienced assessor had a faecal smear sensitivity of 61% and a specificity of 100%. The inexperienced assessor had a faecal smear sensitivity of 54% and a specificity of 95%.

Clinical Significance: Faecal smear is an effective and cheap aid to diagnosis of canine angiostrongylosis which can be readily performed in general practice. Its use leads to a rapid detection of infection compared to other available methods. This could lead to swifter treatment of the disease and a decrease in inappropriate diagnostics. A further diagnostic method, such as Baermann sedimentation, is recommended whether the faecal smear result is positive or negative.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials And Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Angiostrongylus vasorum is a metastrongylid parasite that commonly causes respiratory disease and bleeding diatheses in dogs. Canids (domestic dogs and foxes) are the definitive hosts of A. vasorum with gastropod molluscs (slugs and snails) acting as intermediate hosts (Morgan and others 2005). First-stage larvae (L1) are passed in the faeces of infected dogs after a variable prepatent period, which has been experimentally determined as between 33 and 76 days (Prestwood and others 1981, Oliveira-Junior and others 2006, Barcante and others 2008).

Angiostrongylus vasorum is endemic in many areas in Europe including Southern Britain. Its distribution is spreading with recent case reports in northern England and Scotland (Helm and others 2009, Yamakawa and others 2009). Previous studies have found that dogs under one year of age appear to be at increased risk of disease secondary to A. vasorum (Chapman and others 2004, Koch and Willesen 2009). Staffordshire bull terriers and cavalier King Charles spaniels are reported to be over-represented amongst infected dogs in the UK, although this has not been documented in other countries (Chapman and others 2004, Koch and Willesen 2009).

Definitive diagnosis of A. vasorum is generally made by Baermann faecal sedimentation or bronchoalveolar lavage (BAL); faecal flotation is comparatively less reliable (Traversa and Guglielmini 2008). Polymerase chain reaction (PCR), real-time PCR, Western blot and enzyme-linked immunosorbant assay (ELISA) tests for A. vasorum have been evaluated but are not currently commercially available (Cury and others 2002, Caldeira and others 2003, Verzberger-Epshtein and others 2008, Denk and others 2009, Helm and others 2009, Jefferies and others 2009). As infected animals can be severely ill, rapid diagnosis of A. vasorum would be beneficial but many of the currently available tests take one to two days before results are available. In-house faecal smear analysis is a rapid, bedside test that can be used to support a diagnosis of angiostrongylosis but is not well described in the literature (Patteson and others 1993, Willesen and others 2008). The aim of this study is to describe a method of faecal smear production suitable for use in practice and assess its sensitivity and specificity compared to Baermann faecal sedimentation for the detection of parasitic larvae.

Materials And Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials And Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Faecal samples submitted to the Royal Veterinary College Diagnostic Laboratory Service (RVC DLS) for Baermann sedimentation for detection of A. vasorum over a 15-month period were analysed. Baermann sedimentation was performed on all samples using a previously described standard technique (Zajac 1994). Samples were deemed positive if any A. vasorum larvae were present. The number of larvae seen per gram of faeces analysed was recorded.

Residual faeces were used to produce a faecal smear. Smears were produced in a standardised manner; a lentil-sized volume of faeces was mixed with one drop of tap water on a slide using a gloved finger. The smear was then examined microscopically using the 10× objective with the condenser lowered. Each smear was examined by an experienced and an inexperienced assessor. The experienced assessor had examined approximately 50 faecal smears for A. vasorum before the initiation of this study. The inexperienced assessors were final year veterinary students with little previous experience of faecal smear examination. A different student examined each faecal smear. The students were shown photographs of an A. vasorum larva and a hair as seen microscopically and video of a live A. vasorum larva before examining the faecal smear.

The faecal smears were examined by the assessors for up to five minutes. If the assessor decided they had seen a parasitic larva the smear was classed as positive. If the sample was deemed to be positive during the five-minute assessment period, the time taken to reach this decision was recorded. If this did not occur the assessor stated whether the sample was thought to be positive or negative at the end of the five-minute period.

Using the Baermann sedimentation results, faecal smear sensitivity and specificity values were calculated. The positive predictive and negative predictive values were also calculated. Predictive values provide information about the likelihood that a dog is or is not affected by angiostrongylosis given the test result (Petrie 2000). The number of larvae per gram of faeces found using the Baermann technique was compared to the likelihood of detecting larvae with faecal smear.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials And Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

One hundred and eighty-six faecal samples were submitted for Baermann sedimentation to the RVC DLS during the period of the study. Twenty-eight of these samples (15%) were positive for A. vasorum. No other metastrongylid larvae were identified during the period of this study. The number of larvae per gram of faeces in the positive samples ranged from 0·13 to greater than 1000 larvae per gram.

Seventeen of the 28 Baermann-positive samples were deemed positive using faecal smear by the experienced assessor compared to 15 by the inexperienced assessors. None of the 158 Baermann-negative samples were deemed positive by the experienced assessor, but eight were by the inexperienced assessors. The sensitivity of faecal smears examined by the experienced assessor was 61% and the specificity was 100%. For the inexperienced assessors, the sensitivity was 54% and the specificity was 95%. The positive predictive values were 100% for the experienced assessor and 65% for the inexperienced assessor. The negative predictive value for the experienced assessor was 93 and 92% for the inexperienced assessor.

The median time taken to correctly decide a faecal smear was positive was 29 seconds (range one second to two minutes 24 seconds) for the experienced observer and one minute 10 seconds (range 10 seconds to five minutes) for the inexperienced observers. The experienced assessor correctly classified all samples with a larval concentration of greater than or equal to 66 larvae per gram of faeces as positive. The highest concentration of larvae not detected by the experienced assessor was 24 larvae per gram of faeces. In comparison, no cut-off point was identified with the inexperienced assessors. The lowest larval concentration detected by an inexperienced assessor was 16 larvae per gram of faeces and the highest larval concentration not detected was greater than 1000 larvae per gram of faeces.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials And Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

There is no ideal method for the diagnosis of canine angiostrongylosis. Infected dogs can present with a variety of clinical signs including coagulopathy, respiratory or neurological disease. Given the range of differential diagnoses and the potential severity of disease, rapid diagnosis decreases unnecessary investigation, morbidity and mortality and allows an appropriate prognosis to be communicated to the client. Baermann sedimentation is a commonly used method for the diagnosis of A. vasorum. As this is generally performed as an overnight test at an external laboratory (Zajac 1994), there is a two-day lag between obtaining a faecal sample and the Baermann sedimentation result. BAL can be rapidly diagnostic if cytological analysis is performed in house. However, the procedure is not suitable for dogs with severe respiratory compromise and cytological samples are often sent to external laboratories increasing the time to diagnosis. Blood testing, with PCR or ELISA methods, is appealing as the tests are minimally invasive and studies have shown promising results (Verzberger-Epshtein and others 2008, Helm and others 2009). Unfortunately, blood tests are not currently commercially available. Given these difficulties, this study has examined the utility of faecal smear to aid diagnosis of A. vasorum infestation in dogs. It is a minimally invasive test, gives rapid results and is very cheap. It also requires a very small volume of faeces unlike the minimum of 5 g recommended for Baermann sedimentation (Zajac 1994).

This study found a sensitivity of 61% for the experienced assessor and 54% for the inexperienced assessor when faecal smear examination was compared to Baermann sedimentation for identification of parasitic larvae. The concentration of larvae appears important in determining the likelihood of detection for the experienced assessor. All samples with a concentration of larvae greater than or equal to 66 larvae per gram of faeces were classed as positive and all samples with a larval concentration below this were classed as negative. This relationship was not seen for the inexperienced assessors. This is probably because of variability between inexperienced individuals in recognition of larvae and the variable distribution of larvae within the faecal smear. The significance of the increased detection rate of higher larval concentrations is unknown as there is no recognised relationship between faecal larval concentration and severity of clinical signs. Further investigation in this area would be interesting.

The positive predictive value of the experienced assessor was much greater than that of the inexperienced assessor (100% compared to 65%). If the sample was thought to be positive by the experienced assessor, it was always found to be so with Baermann sedimentation. A positive result from the inexperienced assessor, however, was incorrect 35% of the time. Consequently, care should be taken when interpreting a positive faecal smear result if the assessor is inexperienced. The negative predictive values were high for both experienced and inexperienced assessors at 93 and 92%, respectively. Therefore, if a smear was deemed to be negative on faecal smear it was likely to be negative on Baermann. The reason for the high negative predictive value was the low prevalence of A. vasorum in the samples analysed. If the prevalence was higher the positive and negative predictive values would change, although sensitivity and specificity would remain the same (Petrie 2000). The risk of a false positive in faecal smears being evaluated by inexperienced assessors was important in this study. One inexperienced assessor gave a positive faecal smear result when the concentration of larvae in the sample was 24 larvae per gram of faeces (below the experienced assessor’s detection level). This case was identified to be a false positive (a misidentified hair) with a coincident positive Baermann.

It should be noted that in this study A. vasorum was not definitively identified by faecal smear, merely the presence of parasitic larvae. Other L1 larvae can be present in canine faeces including Crenasoma vulpis, Oslerus osleri and Filaroides species (Traversa and Guglielmini 2008). At the RVC DLS, species-specific diagnosis of larvae is performed by microscopic evaluation of the larvae found on Baermann sedimentation using the ×40 objective and looking for the characteristic signs of A. vasorum larvae, that is a curved tail and a dorsal cuticular spine (Prestwood and others 1981, Bolt and others 1994, Traversa and Guglielmini 2008, McGarry and Morgan 2009). The aim of the faecal smear is rapid diagnosis in a practice setting and we do not suggest species-specific diagnosis should be attempted. If parasitic larvae are seen in conjunction with appropriate clinical signs, a strong suspicion of angiostrongylosis is justified. It is noteworthy that no L1 species other than A. vasorum were detected in the samples in this study by the RVC DLS.

The use of the Baermann sedimentation method as a “gold standard” in this study could be questioned. It is known that faecal larval concentration can vary markedly on a temporal basis (Oliveira-Junior and others 2006) and this has led some to suggest that Baermann sedimentation should be performed on faecal samples collected on three consecutive days to increase sensitivity (Koch and Willesen 2009). ELISA analysis has been shown to be superior to Baermann sedimentation for diagnosis of dogs with clinical disease (Verzberger-Epshtein and others 2008). This is probably, at least in part, because of the fact that the prepatent period (33 to 76 days) needs to be complete before larvae are present in the faeces. Given the common use of Baermann sedimentation for diagnosis of angiostrongylosis and given the lack of true commercially available gold standard, it was felt that for the purposes of this study its use was justified.

Faecal samples in the study were submitted to the RVC DLS and then Baermann sedimentation was performed before faecal smear analysis, meaning the faeces examined were usually several days old. This may have decreased smear sensitivity as live larvae were not commonly seen. In the authors’ experience live larvae are common in fresh faecal smears and their movement makes them easier to detect. Therefore, the sensitivity of fresh faecal smear examination may be higher than that reported in this study. The time limit of five minutes may have also decreased sensitivity as a longer examination period may have led to increased detection of larvae. This was probably less applicable for the experienced assessor as the maximum time taken for declaration of faecal smear positivity was two minutes 24 seconds, less than half the time available. Several inexperienced assessors took five minutes before deciding their faecal smear was positive. A final potential complicating factor is that the study relied upon samples submitted to the RVC DLS. It is possible that faecal smears were performed in practice and only “faecal smear negative” samples submitted for Baermann sedimentation. This again could decrease the faecal smear sensitivity in this study as the proportion of samples with a larval concentration readily detectable by smear analysis would be lower than that in the typical population presenting to the practitioner.

In conclusion, faecal smear analysis can be used to aid diagnosis of A. vasorum infection. The sensitivity of the technique is fair and the specificity is good with both improving with experience. Faecal smear analysis to assess for A. vasorum should only be performed if angiostrongylosis is suspected and the animal is showing compatible clinical signs.

If a negative faecal smear result is obtained, another method of detection such as Baermann sedimentation should be performed. If the faecal smear result is positive another method of detection is still recommended to allow definitive species-specific diagnosis of larvae. This is particularly important if the patient is not displaying classical clinical signs or if the assessor is not experienced due to the mediocre positive predictive value of inexperienced assessors identified in this study. Given the minimal risk of adverse effects and the low cost associated with treating a patient for A. vasorum with licensed products, this should be strongly considered in a patient suspected to be suffering from the disease while waiting for confirmatory tests whatever the faecal smear result. The rapid nature of the faecal smear may allow early diagnosis of angiostrongylosis and therefore prompt and appropriate treatment, decreasing the likelihood of unnecessary and potentially costly investigation and medication.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials And Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Many thanks to the Royal Veterinary College Diagnostic Laboratory Services team for their help with sample collection, the many RVC students who took part in the study and the Epidemiology Department for statistical advice.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials And Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Barcante, J. M., Barcante, T. A., Ribeiro, V. M., Oliveira-Junior, S. D., Dias, S. R., Negrao-Correa, D. & Lima, W. S. (2008) Cytological and parasitological analysis of bronchoalveolar lavage fluid for the diagnosis of Angiostrongylus vasorum infection in dogs. Veterinary Parasitology 158, 93102
  • Bolt, G., Monrad, J., Koch, J. & Jensen, A. L. (1994) Canine angiostrongylosis: a review. Veterinary Record 135, 447452
  • Caldeira, R. L., Carvalho, O. S., Mendonca, C. L., Graeff-Teixeira, C., Silva, M. C., Ben, R., Maurer, R., Lima, W. S. & Lenzi, H. L. (2003) Molecular differentiation of Angiostrongylus costaricensis, A. cantonensis, and A. vasorum by polymerase chain reaction-restriction fragment length polymorphism. Memórias do Instituto Oswaldo Cruz 98, 10391043
  • Chapman, P. S., Boag, A. K., Guitian, J. & Boswood, A. (2004) Angiostrongylus vasorum infection in 23 dogs (1999–2002). Journal of Small Animal Practice 45, 435440
  • Cury, M. C., Guimaraes, M. P., Lima, W. S. & Vitor, R. W. (2002) Western blot analysis of the humoral response of dogs experimentally infected with Angiostrongylus vasorum (Baillet, 1866). Veterinary Parasitology 106, 8387
  • Denk, D., Matiasek, K., Just, F. T., Hermanns, W., Baiker, K., Herbach, N., Steinberg, T. & Fischer, A. (2009) Disseminated angiostrongylosis with fatal cerebral haemorrhages in two dogs in Germany: a clinical case study. Veterinary Parasitology 160, 100108
  • Helm, J., Gilleard, J. S., Jackson, M., Redman, E. & Bell, R. (2009) A case of canine Angiostrongylus vasorum in Scotland confirmed by PCR and sequence analysis. Journal of Small Animal Practice 50, 255259
  • Jefferies, R., Morgan, E. R. & Shaw, S. E. (2009) A SYBR green real-time PCR assay for the detection of the nematode Angiostrongylus vasorum in definitive and intermediate hosts. Veterinary Parasitology 166, 112118
  • Koch, J. & Willesen, J. L. (2009) Canine pulmonary angiostrongylosis: an update. Veterinary Journal 179, 348359
  • McGarry, J. W. & Morgan, E. R. (2009) Identification of first-stage larvae of metastrongyles from dogs. Veterinary Record 165, 258261
  • Morgan, E. R., Shaw, S. E., Brennan, S. F., De Waal, T. D., Jones, B. R. & Mulcahy, G. (2005) Angiostrongylus vasorum: a real heartbreaker. Trends in Parasitology 21, 4951
  • Oliveira-Junior, S. D., Barcante, J. M., Barcante, T. A., Dias, S. R. & Lima, W. S. (2006) Larval output of infected and re-infected dogs with Angiostrongylus vasorum (Baillet, 1866) Kamensky, 1905. Veterinary Parasitology 141, 101106
  • Patteson, M. W., Gibbs, C., Wotton, P. R. & Day, M. J. (1993) Angiostrongylus vasorum infection in seven dogs. Veterinary Record 133, 565570
  • Petrie, A. & Sabine C. (2000) Diagnostic tools. In: Medical Statistics at a Glance. Blackwell Science Ltd, Oxford, UK. pp 9092
  • Prestwood, A. K., Greene, C. E., Mahaffey, E. A. & Burgess, D. E. (1981) Experimental canine angiostrongylosis. Journal of the American Animal Hospital Association 17, 491497
  • Traversa, D. & Guglielmini, C. (2008) Feline aelurostrongylosis and canine angiostrongylosis: a challenging diagnosis for two emerging verminous pneumonia infections. Veterinary Parasitology 157, 163174
  • Verzberger-Epshtein, I., Markham, R. J., Sheppard, J. A., Stryhn, H., Whitney, H. & Conboy, G. A. (2008) Serologic detection of Angiostrongylus vasorum infection in dogs. Veterinary Parasitology 151, 5360
  • Willesen, J. L., Bjornvad, C. R. & Koch, J. (2008) Acute haemoabdomen associated with Angiostrongylus vasorum infection in a dog: a case report. Irish Veterinary Journal 61, 591593
  • Yamakawa, Y., McGarry, J. W., Denk, D., Dukes-McEwan, J., Macdonald, N., Mas, A., McConnell, F., Tatton, B., Valentine, E. G., Wayne, J., Williams, J. M. & Hetzel, U. (2009) Emerging canine angiostrongylosis in northern England: five fatal cases. Veterinary Record 164, 149152
  • Zajac, A. M. (1994) Fecal examination in the diagnosis of parasitism. In: Veterinary Clinical Parasitology. 6th edn. Eds R. L. K. M. W.Sloss and A. M.Zajac. Iowa State University Press, Ames, IA, USA. pp 393