Dogs with clinically aggressive grade II or III mast cell tumours. Location: Washington State University Veterinary Teaching Hospital, Pullman, USA.

19 dogs.

• Abdominal ultrasonography was used to assess liver and spleen.
• Cytologic evaluation of liver and/or spleen aspirates was performed in all the patients (19), with 10 being classified as grade II and 9 as grade III.

Retrospective study.

Objective assessment of the specificity and sensitivity of ultrasound for detecting mast cell infiltration and defining the prognostic role of cytologic evaluation of liver and spleen aspirates.

• Sensitivity of ultrasound for detecting mast cell infiltration was 43% for the spleen (3/7) and 0% for the liver (0/3).
• Specificity of ultrasound abnormalities for predicting mast cell infiltration was 75% for the spleen (9/12) and 91% for the liver (10/11).
• The positive predictive value for detection of mast cell infiltration was 50% for the spleen (3/6) and 0% for the liver (0/1).
• The negative predictive value was 69% for the spleen (9/13) and 77% for the liver (10/13).
• Percentages above are derived from cytology-confirmed counts; x/y are shown for transparency.

• There was a lack of histological confirmation for most cases, because cytology was used as the reference standard.
• The sample size was small, which reduces the precision of the accuracy estimates.
• This was a single-centre study, which may limit generalisability.
• Not all dogs had diagnostic liver aspirates, which introduces potential verification bias.
• Ultrasound interpretations may have been operator-dependent, which could introduce review bias.

Dogs with a confirmed diagnosis of multicentric lymphoma. Location: Auburn University College of Veterinary Medicine, Auburn, USA.

28 dogs.

• Abdominal ultrasonography was used to assess liver and spleen.
• Fine-needle aspiration (FNA) samples were obtained from three sites each in both liver and spleen.
• Cytologic evaluation of the liver and/or spleen aspirates was performed.

Prospective study.

Objective assessment of the accuracy of greyscale ultrasonography in detecting hepatic and splenic lymphoma compared to cytology.

• Liver (n = 69; true positives (TP) = 24, false negatives (FN) = 9, false positives (FP) = 7, true negatives (TN) = 29): sensitivity 72.7% (24/33); specificity 80.6% (29/36); positive predictive value (PPV) 77.4% (24/31); negative predictive value (NPV) 76.3% (29/38); accuracy 76.8% (53/69).
• Spleen (n = 72; true positives (TP) = 42, false negatives (FN) = 0, false positives (FP) = 23, true negatives (TN) = 7): sensitivity 100% (42/42); specificity 23.3% (7/30); positive predictive value (PPV) 64.6% (42/65); negative predictive value (NPV) 100% (7/7); accuracy 68.1% (49/72).
• Sampling adequacy: hepatic cytology 82.1% (69/84) diagnostic; splenic cytology 85.7% (72/84) diagnostic.
• Ultrasonographic patterns: in the liver, lymphoma was identified across all reported appearances including normal parenchyma; in the spleen, a “moth eaten” appearance had PPV 100% (10/10 evaluable), and ultrasonographically normal spleens had NPV 100% (7/7).
• Complications: minor complications occurred in 3/30 (10%) sampled patients (two small subcutaneous haematomas; one small volume of free peritoneal fluid); no severe events and all resolved without treatment.

• The small sample size may affect the generalisability of the findings.
• Only newly diagnosed, untreated dogs were included, which may limit applicability to pretreated populations.
• The diagnostic sampling rate was 82.1% (69/84).

Dogs with visible ultrasonographic changes. Location: Faculty of Veterinary Medicine, University of Zagreb, Zagreb, Croatia.

130 dogs.

• Fine-needle fenestration biopsy (FNFB) (non-aspiration technique) of the spleen in were performed in 130 dogs, with only 75 meeting the inclusion criteria.
• As our PICO focuses on dogs with ultrasonographic splenic changes undergoing FNFB, we do not follow up the results from the remaining 55/130 (42.3%) dogs that did not meet these criteria.

Retrospective study.

To assess objectively, in dogs with ultrasonographically visible splenic changes, whether splenic FNFB cytology was clinically relevant or not, and to assess association between specific ultrasound features and those cytology categories.

• Incidence of clinically relevant diagnoses was 37.3% (28/75) with 82.1% (23/28) of malignant diagnoses (42.9% (12/28) lymphoma).
• No correlation was found between ultrasound lesions examined and the cytological diagnoses, except for splenic heterogeneity (patchy echotexture).
• Lesion size > 1.74 cm on ultrasound predicted clinically relevant cytology, having a sensitivity of 91.3% and specificity of 42.1% for predicting clinically relevant findings.

• The procedures were conducted by different practitioners, therefore results will have operator variability in both ultrasonographic assessment and sampling.
• Retrospective design with inherent limitations in data collection and image consistency.
• There was lack of histological evaluation.
• The clinically relevant group was not limited to round cell neoplasia.

Dogs and cats with diffuse liver disease, evaluated using ultrasonography and confirmed with histology or cytology. Location: University of Minnesota Veterinary Medical Center, Saint Paul, USA.

333 patients (229 dogs and 104 cats).

• Retrospective and independent analysis of ultrasonographic images by three radiologists.
• There were eight histologic or cytologic categories: normal, inflammation, round cell neoplasia, non-round-cell infiltrative, pre nodular (early) metastatic neoplasia, lipidosis, vacuolar hepatopathy, and other.
• Applied ultrasonographic criteria included liver echogenicity, echotexture, borders, and edges; portal venous clarity; sound attenuation of liver parenchyma; gallbladder wall thickness and content; bile duct diameter; caudal vena cava status; and hepatic vein diameter.

Retrospective study.

Accuracy of ultrasonographic criteria in distinguishing eight defined categories of diffuse liver disease in dogs and cats.

Ultrasonographic appearance was insufficient to discriminate among canine and feline diffuse infiltrative liver disease, including round cell neoplasia.

• The retrospective design has inherent limitations in data collection and image consistency.
• The study was limited to diffuse changes, excluding the cases with nodular or focal hepatopathy.
• Histologic or necropsic evaluation was not available in every case.
• The subjective ultrasonographic criteria limits the reproducibility of the study.

Dogs with cutaneous mast cell tumours (cMCT) and ultrasonographically normal liver and spleen, and clinically normal dogs. Location: Animal Medical Center, New York, USA; and College of Veterinary Medicine, Colorado State University, Fort Collins, Colorado, USA.

83 dogs.

• The cohort comprised 51 affected dogs (23 grade II and 28 grade III) and 32 clinically normal dogs.
• Fine-needle aspiration (FNA) of the liver and spleen were performed in both groups.
• The cytology findings were compared between affected and unaffected dogs.

Retrospective study.

Evaluation of the utility of routine hepatic and splenic aspirates in the staging of cMCT.

• There was no significant difference in liver aspirates between affected and unaffected dogs.
• The affected dogs had significantly more mast cells per cluster (P = 0.04) and more isolated mast cells per slide (P = 0.03) in the spleen.
• There was considerable overlap in mast cell numbers between both patient groups (normal and dogs with cMCT), making differentiation unreliable (i.e., many indivicual cMCT dogs had counts within the range observed in normal dogs and vice versa).
• This overlap indicates that routine aspiration of an ultrasonographically normal liver and spleen in cMCT staging is not clinically useful.

• The retrospective design limits control over inclusion criteria and data capture.
• Histopathology of liver and/or spleen was not performed in affected dogs, limiting confirmation of cytology.
• Subjective interpretation of mast cell numbers in cytology and absence of clear cutoff values for distinguishing normal from neoplastic mast cell infiltration. This may impair reproducibility.

Dogs with splenic abnormalities detected during abdominal ultrasonography. Location: Vertsuisse Faculty, University of Zurich, Switzerland.

60 dogs.

• Contrast harmonic ultrasonography was used after splenic abnormalities were detected during routine B-mode abdominal ultrasonography.
• The diagnosis was based on ultrasound-guided aspirates or histopathologic samples taken during surgery or necropsy.

Prospective study.

Objective assessment of the ability of contrast harmonic imaging to differentiate benign from malignant splenic lesions.

• The presence of marked enhancement is of limited value as it occurred in both benign and malignant splenic lesions. Therefore, fine needle aspiration or tissue core biopsy is needed.
• Extensive to moderate hypoechogenicity of a splenic lesion during all blood pool phases was highly associated with malignancy.

• Statistical analysis included different types of splenic tumours, not only round cell neoplasia – possible bias during the evaluation of subjective contrast harmonic ultrasound criteria.
• Only cytology was performed in most of the investigated lesions.
• Histology or cytology of the normal splenic tissue of diseased spleen was not performed.

Survey of ultrasound records in dogs and cats with cytologically or histopathologically confirmed abdominal cell disease.

19 patients (12 dogs and 7 cats).

• For the purpose of this Knowledge Summary, the dog's findings only will be followed up on.
• Ultrasound findings were correlated with confirmed mast cell infiltration in the
• Final diagnoses were obtained by fine-needle aspiration, ultrasound-guided core biopsy of the affected organs, laparotomy, or necropsy.

Retrospective study.

Ultrasonographic features of hepatic, splenic, lymphatic, gastrointestinal, and renal mast cell infiltration.

• Ultrasonographic findings in dogs with mast cell infiltration included increased liver size, a diffuse increase in liver echogenicity, and presence of one or more hypoechoic nodules in the liver or spleen.
• Mast cell infiltration was found in 2 ultrasonographically unremarkable livers and one spleen. Hence, a normal ultrasonographic appearance of the liver or spleen does not rule out mast cell infiltration, highlighting the importance of cytology or histopathology.

• The small cohort reduces precision and external validity.
• The retrospective design limits control over imaging protocols, data quality and consistency.
• The study was carried out over 20 years ago and the findings are based on ultrasound equipment at the time, which limits comparability with current practice.

Dogs with cutaneous mast cell tumours (cMCT), with and without evidence of mast cell infiltration in the spleen and liver. Location: University of Milan, Milan, Italy.

52 dogs.

• 10 dogs with mast cell infiltration and 42 without.
• Ultrasound-guided cytology of spleen and liver was performed.
• The cytology findings and survival time were compared in dogs that had positive and negative results.

Retrospective study.

Survival times of dogs with and without cytological evidence of mast cell infiltration in the spleen and liver.

• 19% of dogs (10/52) had cytological evidence of mast cell infiltration in spleen or liver.
• Dogs with positive cytological evidence of mast cell infiltration had significantly shorter survival times (median 34 days vs 733 days, P = 0.0001).
• 70% (7/10) of the dogs with mast cell infiltration had two or more ultrasonographic abnormalities in the spleen and/or liver.
• Cytology of spleen and liver is recommended in staging, regardless of the ultrasonographic appearance.

• The retrospective design of the study limits the statistical power and external validity.
• The absence of histopathological confirmation for cytologic mast cell infiltration may allow misclassification.
• The small sample size in the group with mast cell infiltration.

Dogs and cats with cytologically or histologically confirmed lymphoma. Location: Small Animal Teaching Hospital, University of Liverpool, UK.

161 animals (132 dogs and 29 cats).

• Blinded evaluation of ultrasonographic images of the liver and spleen were taken contemporaneously with the diagnosis by two board-certified veterinary radiologists.
• The diagnosis of lymphoma was confirmed on cytology and/or histopathology.

Blinded retrospective evaluation.

• One aim was to determine the diagnostic accuracy of ultrasonography in the detection of lymphomatous infiltration of the liver and spleen in a population of dogs and cats with lymphoma.
• Another aim was to determine if specific ultrasonographic features of the liver and spleen in dogs are associated with lymphomatous infiltration or a specific immunophenotype of multicentric lymphoma.

• For the liver: sensitivity 15.8% (true positives (TP) 9 / [TP+false negatives (FN) 57] = 9/57); specificity 91.1% (true negatives (TN) 41 / [TN+false positives (FP) 45] = 41/45); positive predictive value (PPV) 69.2% (TP/(TP+FP) = 9/13); negative predictive value (NPV) 46.1% (TN/(TN+FN) = 41/89); accuracy 49.0% ((TP+TN)/total = 50/102).
• For the spleen: sensitivity 77.5% (TP 55 / [TP+FN 71] = 55/71); specificity 91.5% (TN 43 / [TN+FP 47] = 43/47); PPV 93.2% (TP/(TP+FP) = 55/59); NPV 72.9% (TN/(TN+FN) = 43/59); accuracy 83.1% ((TP+TN)/total = 98/118).
• Hence, ultrasonography of the spleen and liver is specific but not sensitive in the detection of lymphomatous infiltration.
• In dogs, an ultrasonographically normal liver was associated with not having lymphomatous
• A leopard-spotted splenic parenchyma in dogs is highly specific for lymphomatous infiltration and predicted a specific immunophenotype of multicentric lymphoma.

• The retrospective design and review of still images limit standardisation and may introduce selection and information bias.
• Multiple operators and different ultrasound machines/probes may increase variability.
• Both radiologists were trained at the same institution, which may have introduced bias in the interpretation of images and interobserver agreement.
• Using cytology and histology as mixed reference standards may introduce verification bias.

Dogs that had abdominal ultrasonography and abnormal liver findings on biopsy or necropsy. Location: Queen Mother Hospital for Animals, Royal Veterinary College, North Mymms, UK.

371 dogs.

Abdominal ultrasound followed by biopsy or necropsy for histological diagnosis.

Retrospective review.

Correlating ultrasonographic features and histological analysis of abnormal liver.

• No statistically significant associations were found between specific ultrasonographic features and histological diagnoses.
• Ultrasonography is an insensitive test and tentative diagnoses should not be performed based on ultrasonographic findings alone. Histological examination remains essential for accurate diagnosis of liver disease.

• The retrospective design with multiple observers may introduce variability and lesion-matching imprecision.
• The focus on the liver limits the ability to generalise to other abdominal organs.
• The differences between imaging and sampling locations may have affected concordance.

Dogs with primary and metastatic hepatic neoplasia. Location: University of Minnesota, Veterinary Teaching Hospital, Saint Paul, USA.

48 dogs.

• Ultrasonographic findings in dogs with histological confirmation.
• The analysis was focused on the correlation between sonographic appearance and neoplastic cell types.

Retrospective study.

Ultrasonographic patterns associated with different types of hepatic neoplasms, including primary and metastatic tumours.

• Focal hyperechoic masses were frequently hepatocellular carcinomas (14/15 dogs).
• Two patterns were noted for lymphosarcoma: diffuse mildly hyperechoic (6/11) and multifocal hypoechoic (5/11).
• No specific sonographic pattern could reliably predict the neoplastic cell type for focal or multifocal hypoechoic lesions.
• Subtle sonographic changes in the presence of elevated liver enzymes should prompt a biopsy to distinguish between neoplastic and non-neoplastic processes.

• Retrospective design and small numbers with tumour subtypes limit statistical analysis and external validity.
• The study was carried out over 20 years ago and the findings are based on ultrasound equipment at the time, which limits comparability with current practice.

Dogs with histologically confirmed splenic lymphosarcoma. Location: Colorado State University, Fort Collins, USA.

12 dogs.

• Ultrasonographic features in dogs diagnosed with splenic lymphosarcoma were reviewed.
• The spleen and adjacent organs were assessed.

Retrospective study.

Ultrasonographic patterns observed in splenic lymphosarcoma including nodule characteristics and spleen echogenicity.

• All dogs exhibited poorly marginated hypoechoic to anechoic nodules ranging from 4 mm to 3 cm in diameter.
• A hypoechoic spleen was noted in 9/12 dogs when compared with the liver or renal cortices.
• Recognition of these patterns can be a useful diagnostic tool for splenic lymphosarcoma, and ultrasonographically guided needle aspiration will help confirming the diagnosis histologically.

• The retrospective design limits control over image acquisition, quality and consistency.
• The small sample size reduces precision and external validity.
• The study was carried out over 20 years ago and the findings are based on ultrasound equipment at the time, which limits comparability with current practice.

CAB Abstracts on OVID Platform (1973–Week 9 2025)
PubMed accessed via the NCBI website (1920–1 March 2025)

CAB Abstracts:

1. (dog or dogs or canine or canines or bitch or bitches).mp. or exp dogs/ or exp bitches/
2. ((liver* or hepatic* or spleen* or splenic*).mp. or exp liver/ or exp spleen/) and (round cell neoplas* or round cell tumour* or round cell tumor* or lymphoma* or lymphosarcoma* or mast cell tumour* or mast cell tumor* or MCT or histiocytic sarcoma or plasma cell tumour or plasma cell tumor or multiple myeloma).mp.
3. (ultrasound or ultrasonograph* or sonograph*).mp. or exp ultrasonography/ or exp ultrasound/ or exp ultrasonic diagnosis/
4. (cytolog* or histolog* or histopatholog* or biopsy).mp. or exp cytology/ or exp histopathology/
5. 1 and 2 and 3 and 4

PubMed:

1. dog OR canine
2. (liver OR hepatic OR spleen OR splenic) AND ('round cell neoplasia' OR 'round cell tumour' OR 'round cell tumor' OR lymphoma OR lymphosarcoma OR 'mast cell tumour' OR 'mast cell tumor' OR MCT OR ‘histiocytic sarcoma’ OR ‘plasma cell tumour’ OR ‘plasma cell tumor’ OR ‘multiple myeloma’)
3. ultrasound OR sonography OR ultrasonography
4. histology OR histopathology OR biopsy OR cytology
5. 1 AND 2 AND 3 AND 4

1 Mar 2025

Does not answer the PICO question, non-English language, popular press (non-scientific or peer-reviewed), human medicine literature.

Any primary veterinary research/systematic review that compared the accuracy of ultrasound compared to cytology / histopathology.