Background: Although B-mode ultrasound is very sensitive for the detection of splenic lesions, its specificity is low. Contrast harmonic imaging is used successfully to differentiate benign from malignant liver lesions in humans and dogs.Hypothesis: Contrast harmonic imaging could be useful to differentiate benign and malignant splenic lesions in dogs. Animals: Sixty dogs (clinical patients) with splenic abnormalities detected during abdominal ultrasonography. Methods: A prospective study was performed with a Philips ATL 5000 unit for contrast pulse inversion harmonic imaging (mechanical index: 0.08, contrast medium: SonoVue). Perfusion was assessed subjectively and quantitatively.Results: Cytology or histology identified 27 benign (hyperplasia, extramedullary hematopoiesis, hematoma) and 29 malignant (hemangiosarcoma, malignant lymphoma, malignant histiocytosis, mesenchymal tumors without classification, mast cell tumors, and others) lesions and 4 normal spleens. Except for 1 benign nodule, extensive to moderate hypoechogenicity was only seen in malignant lesions during wash-in, at peak enhancement, and during wash-out (P 5 .0001, odds ratios: 37.9 [95% CI 4.5-316.5], 66.4 [95% CI 8.0-551.1], and 36.9 [95% CI 4.4-308.4]). Although all but 1 benign lesion enhanced well and were mildly hypo-, iso-, or hyperechoic in comparison with the normal spleen during all blood pool phases, marked enhancement occurred both in benign as well as in malignant splenic lesions. Quantitative perfusion values did not differ significantly between benign and malignant lesions.Conclusions and Clinical Importance: Moderate to extensive hypoechogenicity clearly identifies canine splenic malignant lesions. In nodules with marked enhancement, contrast harmonic ultrasound is of limited value and histology is needed.

The purpose of this study was to assess the perfusion pattern and perfusion dynamics in the normal canine spleen using contrast harmonic imaging. Twenty-five dogs without clinical or ultrasonographic evidence of splenic disease were studied. Twenty-three dogs were scanned with only manual restraint; two dogs were sedated with buprenorphin. All dogs received an intravenous bolus of a microbubble contrast medium (SonoVue). The perfusion pattern during the blood pool phase represented a skewed bell-shaped curve. A tissue-specific late phase, similar to humans, was not observed. Time/intensity curves were generated for a selected region. Mean average-derived peak intensity (PI) was 6.6dB, mean time to peak intensity calculated from the initial rise (TTP) was 25.6 s and mean area under the curve (AUC) was 523.6 dBs. If dogs were divided into two body weight groups (< or =15 and >15 kg body weight), average derived peak intensity area, time to peak intensity, and area under the curve were lower for the smaller dogs than for the larger animals. However, differences were not statistically significant (P = 0.2, 0.05, and 0.08, respectively). No significant association was found between hematocrit, hemoglobin concentration, red blood cell count, blood pressure, heart rate, age, gender, and the perfusion variables. In conclusion, these baseline data may prove useful in the evaluation of dogs with diffuse or focal splenic disease.