ttp://www.bsava.

com PAPER

Association between ultrasonographic

appearance of urine and urinalysis in
dogs and cats
F. Valls Sanchez1,*, J. Puig*, G. Olivares*, A. Holloway† and M. Seth*

*Department of Small Animal Internal Medicine, Animal Health Trust, Newmarket, Suffolk, CB8 7UU, UK
†
Department of Radiology, Animal Health Trust, Newmarket, Suffolk, CB8 7UU, UK
1
Corresponding author email: [email protected]

Objective: To evaluate association between ultrasonographic urine echogenicity and sediment examina-
tion in dogs and cats.
Materials and Methods: Dogs and cats undergoing ultrasound-guided cystocentesis at a multidisciplinary
referral hospital. Ultrasonographic images were stored and reviewed by a single, blinded, board-
certified radiologist. Urine appearance was described as “echoic” or “anechoic”. Urine sediment was
examined for bacteriuria, pyuria, haematuria, crystalluria and urine-specific gravity and then classified
as “active” or “inactive.”
Results: Of the 194 cases included in this study, urine was echoic in 52 and anechoic in 142. Sediment
was active in 52 and inactive in 142 samples. Sensitivity and specificity of echoic urine for active
sediment were 40% (95% CI: 27 to 55%) and 78% (95% CI: 70 to 85%), respectively. Positive predic-
tive value and negative predictive value of echoic urine for active sediment were 40% (CI 30 to 52%)
and 78% (CI 74 to 82%), respectively. If urine-specific gravity was <1.015 urine was always described
ultrasonographically as anechoic.
Clinical Significance: Association between sediment analysis and ultrasonographic appearance of urine
is poor. Echoic urine had low positive predictive value for active sediment in this study, suggesting
that echoic urine alone should not prompt urinary investigations in the absence of other clinical suspi-
cion. Despite a negative predictive value of 78%, urinalysis is still indicated for anechoic urine, espe-
cially if urine specific gravity is low.

Journal of Small Animal Practice (2019)

DOI: 10.1111/jsap.12996
Accepted: 27 December 2018

INTRODUCTION informative in diagnosis of haematuria, urolithiasis and suspected

neoplastic conditions of the urinary tract. However, crystals in
Urinary sediment analysis is commonly performed in suspected the urinary sediment may differ from the composition of uroliths
lower urinary tract disease in dogs and cats. Urinary sediment (Chew et al. 2011, Lulich & Osborne 2017) and signalment,
examination and bacterial culture are particularly useful for the radiographic appearance and urine pH are considered the main
diagnosis, treatment and monitoring of urinary tract infections tools to predict urolith composition (Lulich & Osborne 2017).
(UTIs). This is a common condition in small animal patients, Abdominal ultrasound may detect changes in the echogenicity
with 14% of all dogs suffering from a UTI during their lifetime of urine but uncertainty over its significance often prompts ultra-
(Pressler & Bartges 2010). Urinary sediment evaluation is also sound-guided cystocentesis and examination of urinary sediment
and bacterial culture. Cystocentesis without clinical indication is
undesirable because it may be associated with a greater need for
F.V. Sanchez current address is Dick White Referrals, Station Farm, London Road, Six Mile
Bottom, Cambridgeshire CB8 0UH sedation, potential procedural complications, increased cost and

Journal of Small Animal Practice • © 2019 British Small Animal Veterinary Association 1
 F. Valls Sanchez et al.

increased personnel time. Although infrequent, complications

can occur (Specht et al. 2002, Buckley et al. 2009, Odunayo et al.
2015), including aspiration of intestinal tract contents, transient
haematuria and haemorrhage from the urinary bladder mucosa
(Buckley et al. 2009). Iatrogenic aortic laceration has also been
reported in a dog, for which massive transfusion and surgery were
required (Buckley et al. 2009). Septic peritonitis following cys-
tocentesis in a dog has also been reported (Specht et al. 2002).
The objective of this study was to evaluate the association
between ultrasonographically echoic urine and the results of uri-
nary sediment examination in dogs and cats.

MATERIALS AND METHODS

FIG 2. Echoic ultrasonographic appearance of the urine
This was a prospective study at a single referral hospital. Any dog
or cat undergoing ultrasound-guided cystocentesis between April
2015 and July 2018 was included. Inclusion criteria were the avail- images were further subcategorised into “particulate/suspended
ability of ultrasonographic images for review and the record of uri- sediment without reverberation,” “particulate/suspended sedi-
nary sediment results. The urinary sediment examination had to ment with reverberation,” “dependent sediment without acoustic
be performed within the first 6 hours after sampling for the case to shadowing” or “dependent sediment with acoustic shadowing.”
be included in the study. Ultrasonographic images were obtained A urine sample was obtained by ultrasound-guided cystocen-
by a board-certified veterinary radiologist or by a resident under tesis at the time of the ultrasound examination. The urine sample
the supervision of a board-certified radiologist using an iU22 was placed in a sterile universal container and submitted to an
ultrasound system (Philips Medical Systems) and a curved (8 to on-site ISO-accredited laboratory. The urine sample was mixed
5 MHz) transducer. A standardised protocol was used in all cases. by gentle inversion and poured into a 20-mL conical tube and
The patient was placed in lateral or dorsal recumbency, depend- centrifuged at 500 to 1000 rpm for 5 minutes. The supernatant
ing on the radiologist’s preference. The urinary bladder had to be was discarded, leaving 0.5 mL in the tube, before the sediment
moderately distended according to the radiologist’s criteria. A sin- was resuspended by aspirating the sample multiple times. One
gle focal zone was in the level of the far wall of the urinary bladder. drop of the suspension was placed on a microscope slide and cov-
The images were obtained through the linea alba to avoid the near- ered with a coverslip without staining. The evaluation was per-
field artefacts produced by the abdominal muscles and optimised formed within 6 hours of after sampling by a trained technician
to minimise any artefact produced by the adjacent descending or a board-certified veterinary clinical pathologist. Erythrocytes
colon. Sagittal and transverse images of the urinary bladder were and leucocytes were counted and the average number of cells
obtained for each case. Recorded DICOM images of the urinary was reported after evaluating 10 fields at ×100 magnification.
bladder were reviewed by a single board-certified veterinary radi- Pyuria and haematuria were diagnosed if there was an average of
ologist blinded to patient signalment, history and urinalysis find- at least five cells per high-powered field (hpf ). Bacteria and crys-
ings. The images were classified as either “anechoic” or “echoic” tals were counted semi-quantitatively: “none/few” (0 to 5/hpf ),
urine (Fig. 1 and 2). If the urine was classified as “echoic”, the “occasional” (5 to 20/hpf ), “moderate” (20 to 100/hpf ), “many”
(>100/hpf ). Active sediment was defined by bacteriuria, pyuria
or haematuria. Urine-specific gravity (USG) was also measured
using a calibrated refractometer and reported for all samples.
Data regarding the signalment (species, age, sex, breed), results
of the urinary sediment examination, bacterial culture result (if
available), USG and type of urine ultrasonographic echogenicity
(echoic or anechoic) were collected and recorded. The cases were
also subclassified depending on the species and whether there
were lower urinary tract signs (dysuria, pollakiuria, strangu-
ria, macroscopic haematuria, urinary incontinence, nocturia or
abnormal odour of the urine according to the owner).
SPSS Statistics for Macintosh version 25.0 (IBM) was used
for statistical analysis. Statistical analysis was performed using the
chi-squared 2×2 contingency test for the relation between the
echogenicity of the urine and the presence of an active sediment,
bacteriuria, haematuria, pyuria and crystalluria. An interim power
FIG 1. Anechoic ultrasonographic appearance of the urine analysis was conducted after 100 samples had been collected

2 Journal of Small Animal Practice • © 2019 British Small Animal Veterinary Association
 Association urinalysis – urine echogenicity

and, based on this, it was calculated that 228 samples would be Table 1. Population characteristics
required to identify a significant difference in the proportion of Sample population (194 cases)
cases with active sediment between echoic and anechoic urine
Canine (140/194)
with an alpha of 0.05 and a power of 80%. Sensitivity, specific- Sex Male neutered 52 of 140 (37%)
ity, positive predictive value and negative predictive value were Female spayed 49 of 140 (35%)
calculated and reported with 95% confidence intervals (CI) for Male entire 23 of 140 (16%)
Female entire 16 of 140 (11%)
the echogenicity of the urine as a tool for diagnosis of active sedi- Breed Labrador retriever: seven of 140 (10%)
ment, bacteriuria, pyuria haematuria and crystalluria. Regarding Cocker spaniel: 11 of 140 (8%)
the USG and its relationship with the echogenicity of the urine, Golden retriever: seven of 140 (5%)
German shepherd dog: seven of 140 (5%)
a Kolmogorov-Smirnov test was used to assess the normality and Crossbreeds 12 of 140 (9%)
a t-test was applied because the data were normally distributed. P 96 individuals from further 46 breeds 96 of 140
values <0.05 were considered to be significant. (63%) with ≤6 individuals each.
Mean age 6.4 years (4 months to 14 years).
Feline (54 of 194)
Sex Male neutered 32 of 54 (59%)
RESULTS Female spayed 19 of 54 (35%)
Male entire one of 54 (2%)
Two hundred and sixty-three cases were initially considered Female entire two of 54 (4%)
Breed Domestic short hair 36 of 54 (67%)
for inclusion in the study. Sixty-nine cases were excluded due 18 individuals from further 11 breeds 18 of 54
to the lack of saved images of diagnostic quality for subsequent (33%) with ≤3 individuals each.
review or missing laboratory data, leaving 194 cases. Regarding Mean age 10.6 years (1 year to 18 years)
the diagnostic quality, images were considered non-diagnostic if
near-field artefact or side-lobe artefact from the colon obscured a
part of the bladder or if one of the two planes was not included Table 2. Distribution of urinary sediment abnormalities
(transverse or sagittal). Table 1 shows the population characteris- from 194 urine samples in dogs and cats
tics for the cases included in the study. Type of urinary sediment abnormality Number of cases
Eighteen of 194 (9%) cases had lower urinary tract signs and Cases with only one abnormality 46
176 of 194 (90%) did not. No complication associated with Bacteriuria 33
cystocentesis was reported in any case. Ultrasonographic appear- Haematuria 7
Crystaluria 6
ance of the urine was considered echoic in 52 of 194 (27%) Cases with more than one abnormality 12
of the cases and anechoic in the remaining 142 of 194 (73%). Bacteriuria+pyuria 2
The echoic images were subcategorised into particulate/sus- Bacteriuria+haematuria 1
Bacteriuria+crystaluria 5
pended sediment without reverberation (45 of 52), particulate/ Bacteriuria+pyuria+haematuria 2
suspended sediment with reverberation (three of 52), dependent Bacteriuria+haematuria+crystaluria 1
sediment without acoustic shadowing (two of 52) and dependent Pyuria+crystaluria 1
Pyuria+haematuria 0
sediment with acoustic shadowing (two of 52). Urine sediment
findings are summarised in Table 2. Tables 3 and 4 show the
distribution of the urine echogenicity depending on the urinary
Table 3. Distribution of echogenicity in relation to the type
sediment parameters and the sensitivity, specificity, positive pre- of sediment in the overall population (active/inactive)
dictive value and negative predictive value of the echogenicity of
Inactive sediment Active sediment Total
the urine for the different results of urinary sediment examina-
Anechoic 111 31 142
tion in the overall population. Echoic 31 21 52
Bacterial culture was performed in 130 of 194 (67%) cases. Total 142 52 194
Using these data, sensitivity and specificity of the urinary sediment Sensitivity 40% (CI 27 to 55%)
detection of bacteriuria in relation to the results of the bacterial Specificity 78% (CI 70 to 85%).
Positive predictive value 40% (CI 30 to 52%).
culture (when available) were calculated (Table 5). Table 6 shows Negative predictive value 78% (CI 74 to 82%).
the distribution of the urine echogenicity in relation to active
sediment and the sensitivity, specificity, positive predictive value
and negative predictive value of the echogenicity of the urine for
active sediment in the population with and without lower urinary urine echogenicity in relation to active sediment and the sensitiv-
tract signs. There was a significant difference in the USG between ity, specificity, positive predictive value and negative predictive
echoic and anechoic samples when considering the overall popu- value of the echogenicity of the urine for active sediment in the
lation (P≤0.001) and in dogs (P=0.034) but not in cats (Fig. 3). canine population.
Within the dogs, ultrasonographic appearance of the urine Within the cats, ultrasonographic appearance of the urine was
was considered echoic in 27 of 140 (19%), active sediment was considered echoic in 25 of 54 (46%), active sediment was diag-
diagnosed in 37 of 140 (26%) and lower urinary tract signs were nosed in 15 of 54 (28%) (Tables 6 and 7). Lower urinary tract
present in 14 of 140 (10%). Table 7 shows the distribution of the signs were present in four of 54 (7.4%). Table 8 shows the dis-

Journal of Small Animal Practice • © 2019 British Small Animal Veterinary Association 3
 F. Valls Sanchez et al.

Table 4. Distribution of echogenicity of urine in relation to bacteriuria, pyuria, haematuria and crystalluria in the overall
population
Bacteriuria Pyuria Haematuria Crystalluria
Yes No Yes No Yes No Yes No
Anechoic 24 117 3 138 5 136 6 135
Echoic 18 34 2 50 6 46 7 45
Total 42 151 5 188 11 182 13 180
Sensitivity 43% (CI 28 to 59%) 40% (CI 5 to 85%) 55% (CI 23 to 83%) 54% (CI 25 to 81%)
Specificity 77% (CI 70 to 84%) 73% (CI 66 to 80%) 75% (CI 68 to 81% 75% (CI 68 to 81%)
Positive predictive value 35% (CI 25 to 46%) 4% (CI 1 to 11%) 12% (CI 7 to 19%) 13% (CI 8 to 21%)
Negative predictive value 83% (CI 79 to 87%) 97% (CI 96 to 99%) 96% (CI 93 to 98%) 96% (CI 93 to 98%)

Table 5. Distribution of bacterial culture results in relation Table 7. Distribution of echogenicity in relation to the type
to bacteriuria in the urinary sediment and the sensitivity of sediment in the dogs (active/inactive)
and specificity of urinary sediment for detection of Inactive sediment Active sediment Total
bacteriuria
Anechoic 90 23 103
Bacterial culture 130/194 (67%) Echoic 13 14 27
Positive Negative Total 103 37 140
Bacteriuria 5 23 28 Sensitivity 38% (CI 22 to 55%)
No bacteriuria 5 97 102 Specificity 87% (CI 79 to 93%)
10 120 130 Positive predictive value 52% (CI 36 to 67%)
Sensitivity of urinary sediment 50% (CI 19 to 81%) Negative predictive value 80% (CI 75 to 84%)
Specificity of urinary sediment 81% (CI 73 to 87%)
Bacteriuria without bacterial culture available 16 of 44 (36%)

Table 6. Distribution of echogenicity in relation to the Table 8. Distribution of echogenicity in relation to the type
type of sediment in the population with and without lower of sediment in the cats (active/inactive)
urinary tract signs (active/inactive) Inactive sediment Active sediment Total
Lower urinary tract signs No lower urinary tract signs Anechoic 21 8 29
Echoic 18 7 25
Sediment Inactive Active Total Inactive Active Total
Total 39 15 54
Anechoic 11 2 13 100 29 138 Sensitivity 47% (CI 21 to 73%)
Echoic 3 2 5 28 19 47 Specificity 54% (CI 37 to 70%)
Total 14 4 18 128 48 176 Positive predictive value 28% (CI 17 to 42%)
Sensitivity 50% (CI 7 to 93%) 40% (CI 26 to 55%) Negative predictive value 72% (CI 60 to 82%)
Specificity 79% (49 to 95%) 78% (CI 70 to 85%)
Positive 40% (14 to 73%) 40% (CI 30 to 52%)
predictive value
Negative 85% (67 to 94%) 78% (CI 73 to 82%)
predictive value DISCUSSION

Our results show that the negative predictive value of urine echo-
genicity was moderate for active sediment (78%, CI 74 to 92%)
and bacteriuria (82%, CI 79 to 87%) and high for pyuria (97%,
CI 93 to 98%) and haematuria (96%, CI 93 to 98%) in the
overall population. Based on these observations, anechoic urine
in cats and dogs is more likely associated with inactive sediment.
This was also true when dogs and cats were analysed separately
(slightly lower negative predictive value in cats – 72% versus
80%). This information can be used as a guide for prioritising
other investigations if the clinical complaint is vague (such as
lethargy or pyrexia), there are no lower urinary tract signs and the
urine is ultrasonographically anechoic. However, future studies in
FIG 3. Plot of the USG in the anechoic and echoic groups
this area may be warranted to determine if the ultrasonographic
appearance of urine can be useful as a diagnostic tool for urinary
tribution of the urine echogenicity in relation to active sediment sediment abnormalities and monitoring tool for such cases and it
and the sensitivity, specificity, positive predictive value and nega- must be remembered that positive and negative predictive values
tive predictive value of the echogenicity of the urine for active are highly dependent on the specific examined population.
sediment in the cats. There was a significant relationship between crystalluria and the
echogenicity of the urine in this study population. The negative

4 Journal of Small Animal Practice • © 2019 British Small Animal Veterinary Association
 Association urinalysis – urine echogenicity

predictive value was high (96%, CI 93 to 98%), suggesting an referral hospital and so the conclusions of this study are not repre-
association of anechoic urine and absence of crystalluria. This sentative for the full population of dogs and cats. Specially, there
information can be useful for monitoring purposes in some clinical was likely to be a higher likelihood of urinary abnormalities and
scenarios such as dogs with history of urolithiasis or on medications this could have had an impact on the positive and negative pre-
reported to promote the formation of calculi such as allopurinol. dictive values. While the ultrasonographic images were acquired
While the negative predictive values for various findings iden- in a standardised manner and scored by a single radiologist, mul-
tified in this study were high, the sensitivities/specificities were tiple radiologists acquired the images and static pictures were used
not. This suggests that the findings are in part due to a low prev- rather than cine loops due to logistical considerations. This may
alence of sediment abnormalities within the study population, limit accurate recognition of echoic urine in some cases and cer-
which were cats and dogs undergoing abdominal ultrasound and tainly precluded any qualitative assessment of the nature of that
cystocentesis at a referral hospital for a variety of indications. echogenicity which may have been more specifically associated
These findings must therefore be interpreted with caution and with certain urine characteristics. Another limitation of the ultraso-
considered with respect to the specific patient population that is nographic evaluation is that no information about urinary bladder
being investigated in a different clinical environment. abnormalities or urinary bladder wall thickness was collected. This
Interestingly, there was a significant USG difference between information could have been used to subcategorise the cases with
the anechoic and echoic group in the population as a whole and a possible impact on the positive and negative predictive values.
in dogs alone. In the current data set, urine samples with a USG Bacteriuria was assessed using unstained sediment examination,
below 1.015 were always anechoic. This information is impor- which may be controversial. One study, using bacterial culture as
tant because it may preclude echogenicity as a method to rule out a gold standard, compared routine unstained preparations and
active sediment in animals with low urine-concentrating abil- modified Wright-stained preparations and concluded sensitivities
ity. This information should be taken into consideration when of 82.4 and 93.2%, and specificities of 76.4 and 99.0%, respec-
approaching the interpretation of urine echogenicity in polyuric tively (the study used quantitative bacteriologic culture as a gold
animals. For example, in animals with chronic kidney disease, in standard) (Swenson et al. 2004). In comparison, another study
which isosthenuric urine is commonly encountered, the echo- compared unstained and stained (Wright Giemsa, Gram) samples
genicity of the urine should not be used to be more or less suspi- for which the sensitivity was 89 and 83% and the specificity of 91
cious about an active sediment. and 99%, respectively (O’Neil et al. 2013). In the present study,
Due to the small number of cases with an echoic urine, it is unstained samples were used, because this is the standard tech-
necessary to be cautious about its clinical importance. In addi- nique of the laboratory that processed the samples and we found
tion, there is the possibility that variation in the echogenicity of a sensitivity of 50% and specificity of 81%. The specificity was
the urine is not related (at least, exclusively) to bacteriuria, pyuria, similar to some previously reported values but the sensitivity was
haematuria or crystalluria and there may be other physicochemi- much lower. However, the number of cases with a positive bacte-
cal properties involved. These data suggested a role of the USG in rial culture was small and the CIs were correspondingly wide. Bac-
the determination of the echogenicity of the urine: urine samples terial culture is the gold standard for identifying bacteriuria (Wood
with low USG are most likely anechoic as previously mentioned. 2017) but was not performed in all our cases (only 67%) and so
However, there was a large overlap regarding the echogenicity of we cannot exclude a number of false negative and false positives.
the urine when the USG was higher. In human literature, partic- More than one abnormality in the urinary sediment was reported
ulate echoes on transvaginal ultrasound were considered normal in some of the samples, but insufficient cases were included for
and not related to any sediment abnormality (Wachsberg et al. a meaningful multi-variate analysis. Similarly, the number of
1998). The authors considered that the particulate echoes were cases with different echogenicity characteristics was low, preclud-
related to other components of the urine. ing meaningful analysis of these sub-categories. Additionally, the
In feline medicine, an association between changes in the severity of each one of the urinary sediment abnormalities was not
echogenicity of the urine and lipiduria has been reported (Sislak assessed, which could have had an impact on the echogenicity clas-
et al. 2014). In that study of healthy cats, 90% of the cases had sification system. Finally, the complete clinical data for each case
particulate echoes suspended in the urinary bladder and 66% of was not collected, making the application of these data to specific
the cases had no significant findings on urinalysis other than con- clinical scenarios more challenging. For example, the use of antibi-
taining lipid. The authors concluded that cats with increased dia- otics was not recorded but could have had an impact on the results
cylglycerol in the urine were more likely to have increased echoes of the urinary sediment and bacterial culture.
on ultrasound. Interestingly, the present study revealed the urine In conclusion, anechoic urine in dogs and cats is likely asso-
to be scored as echoic in 19% dogs whereas it was reported as ciated with inactive sediment, particularly in cases without
echoic in 46% cats, suggesting that cats more commonly have lower urinary tract signs. Anechoic urine also is more sugges-
echoic urine than dogs. Of the cats with echoic urine, only four tive of the absence of bacteriuria, haematuria or crystalluria.
of 25 had an active sediment. The higher percentage of cases However, due to the characteristics of our population this
with echoic urine could be associated with lipiduria, although information should always be considered in light of the clinical
this variable was not quantified in the present study. picture and, for example, should not preclude cystocentesis in
This study has multiple limitations. The sample only included cases with lower urinary tract signs. In summary, these asso-
patients undergoing cystocentesis for various medical reasons in a ciations should only be used as a guide by clinicians on the

Journal of Small Animal Practice • © 2019 British Small Animal Veterinary Association 5
 F. Valls Sanchez et al.

need for a cystocentesis in specific clinical settings. Low USG Chew, D. J., DiBartola, S. P. & Schenck, P. A. (2011) Urolithiasis. In: Canine and
Feline Nephrology and Urology. 2nd edn. Eds D. J. Chew, S. P. DiBartola and P. A.
(particularly below 1.015) in dogs is associated with anechoic Schenck. Elsevier Saunders, St Louis, MO, USA. pp 272-305
urine and this information should be taken into consideration Lulich, J. P. & Osborne, C. A. (2017) Lower urinary tract urolithiasis in dogs. In:
Textbook of Veterinary Internal Medicine. 7th edn. Eds S. J. Ettinger and E. C.
when interpreting the urine echogenicity of polyuric animals. Feldman. W. B. Saunders, Philadelphia, PA, USA. pp 1996-2004
Further studies with a larger sample size are needed to evalu- O’Neil, E., Horney, B., Burton, S., et al. (2013) Comparison of wet-mount, Wright-
Giemsa and gram-stained urine sediment for predicting in dogs and cats. Cana-
ate the utility of ultrasound in specific clinical situations and dian Veterinary Journal 54, 1061-1066
to determine and further characterise the association between Odunayo, A., Ng, Z. Y. & Holford, A. L. (2015) Probable vasovagal reaction fol-
lowing cystocentesis in two cats. Journal of Feline Medicine and Surgery Open
urine echogenicity and specific sediment findings or physico- Reports, 1-4
chemical properties of the urine. Pressler, B. & Bartges, J. W. (2010) Urinary tract infections. In: Textbook of Vet-
erinary Internal Medicine. 7th edn. Eds S. J. Ettinger, E. C. Feldman and W. B.
Saunders, Philadelphia, PA, USA. pp 2036-2037
Acknowledgements Sislak, M. D., Spaulding, K. A., Zoran, D. L., et al. (2014) Ultrasonographic charac-
teristics of lipiduria in clinically normal cats. Veterinary Radiology & Ultrasound
The authors thank Department of Imaging at the Animal Health 55, 195-201
Trust for the collection of the ultrasonographic images. Specht, A., Chan, D., O’Toole, T., et al. (2002) Acute staphylococcal peritonitis
following cystocentesis in a dog. Journal of Veterinary Emergency and Critical
Care 12, 183-187
Conflict of interest Swenson, C. L., Boisvert, A. M., Kruger, J. M., et al. (2004) Evaluation of modi-
fied Wright-staining of urine sediment as a method for accurate detection of
No conflicts of interest have been declared. bacteriuria in dogs. Journal of the American Veterinary Medical Association 224,
1282-1289
Wachsberg, R. H., Festa, S., Saaman, P., et al. (1998) Particulate echoes within
References bladder detected with transvaginal sonography: a sign of urinary tract infection?
Buckley, G. J., Aktay, S. A. & Rozanski, E. A. (2009) Massive transfusion and Emergency Radiology 5, 137-139
surgical management of iatrogenic aortic laceration associated with cysto- Wood, M. W. (2017) Lower urinary tract infections. In: Textbook of Veterinary Inter-
centesis in a dog. Journal of the American Veterinary Medical Association 239, nal Medicine. 8th edn. Eds S. J. Ettinger, E. C. Feldman and E. Cotte. W. B.
288-291 Saunders, Philadelphia, PA, USA. pp 1992-1996

6 Journal of Small Animal Practice • © 2019 British Small Animal Veterinary Association