You are currently on the new version of our website. Access the old version .
MDPIMDPI
Search all articles
DairyDairy
Download PDF
  • Article
  • Open Access

21 November 2025

Ultrasonographic Assessment of Glandular Cistern Area in Dairy Cows with Clinical and Subclinical Mastitis: Feasibility, Reliability, and Diagnostic Implications

,
,
,
,
and
1
Department of Veterinary Sciences, via Livornese snc, San Piero a Grado, 56122 Pisa, Italy
2
Centro di Ricerche Agro-Ambientali “E. Avanzi”, University of Pisa, via Vecchia di Marina 6, San Piero a Grado, 56122 Pisa, Italy
*
Author to whom correspondence should be addressed.
Dairy2025, 6(6), 68;https://doi.org/10.3390/dairy6060068
Version Notes
Order Reprints

Abstract

Ultrasonography has been proposed as a complementary tool for the evaluation of udder health, yet limited information exists on its application for measuring the glandular cistern area during mastitis. This study aimed to investigate modifications of the glandular cistern area in clinical (CM) and subclinical mastitis (SCM) compared with contralateral healthy quarters, and to assess the reliability of manual and automated ultrasonographic measurements. A longitudinal study was conducted on 42 Italian Holstein cows, comprising 26 SCM and 16 CM quarters, each paired with contralateral healthy controls. Ultrasound examinations were performed at diagnosis (T0), 24 h (T1), and 5 days (T5). Cisternal areas were measured in transversal and longitudinal sections using both manual and ImageJ-guided methods. Intra- and inter-operator reliability was assessed using Intraclass Correlation Coefficients (ICCs). Statistical analyses included two-way mixed ANOVA with post-hoc Bonferroni correction. Mastitic quarters tended to show smaller cisternal areas compared with contralateral healthy quarters, with significant differences observed between contralateral healthy and CM quarters (p < 0.05), but not between contralateral healthy and SCM or between SCM and CM quarters. Temporal trends differed significantly among groups (interaction effect, p < 0.05). Both manual and automated measurements demonstrated excellent intra- and inter-operator reliability, with ICCs consistently > 0.95 across pathology groups and time points. Ultrasonography of the glandular cistern is a feasible, reproducible, and reliable method under field conditions. Clinical mastitis is associated with a measurable reduction in cisternal area, while changes in subclinical mastitis appear less evident. The high reproducibility of measurements, including automated analysis, supports the use of this technique may contribute to improve the use of ultrasound also for the udder heath.

1. Introduction

Ultrasonography is a non-invasive diagnostic tool that has been broadly implemented for clinical evaluations of both individual cases and herd monitoring [1,2,3,4,5]. Among its different uses, ultrasound has been applied to investigate udder health and milk production in cows [6,7,8,9,10], sheep [11,12], and goats [13]. Compared with conventional diagnostic approaches, such as clinical inspection, bacteriological culture, somatic cell count, or milk conductivity, ultrasonography provides real-time information on both structural and functional features of the mammary gland [14]. In addition, it can be repeated without interfering with milk production, making it particularly attractive for both research and field applications. In particular, ultrasonography of udder and teats enables to the detect several pathological changes: in the teat, it can reveal obstructions, mucosal alterations, or proliferative processes; in the udder parenchyma, it can identify inflammation, edema, abscesses, fibrosis, hematomas, and neoplastic lesions [14]. These conditions are generally recognized as alterations in echogenicity or disruption of the normal parenchymal pattern, sometimes associated with intraluminal material such as debris, pus, or milk stones [9,14,15].
Mastitis, an inflammatory condition of the mammary gland, involves both bacterial infection and inflammatory oxidative stress. It may be caused by a wide range of pathogens, including bacteria, viruses, fungi, and algae, leading to tissue damage [16] and substantial economic losses in the dairy industry [17]. The disease manifests in two forms: clinical mastitis, characterized by visible udder and milk alterations, and subclinical mastitis, which is detectable only through laboratory testing [18]. Subclinical mastitis is far more prevalent, with a frequency 15–40 times higher than the clinical form [19]. Because it often progresses without overt clinical signs, its impact is linked to decreased milk yield, compositional changes, and elevated somatic cell count (SCC) [20]. In routine practice, diagnosis is therefore commonly based on indirect markers, particularly SCC, with a threshold of 200,000 cells/mL generally applied to differentiate healthy from affected quarters [21]. However, these approaches have limitations, as they either require laboratory support and fail to provide information on the structural changes occurring within the udder. Ultrasonography, in contrast, offers the opportunity to visualize morphological and echogenic alterations in real time, potentially bridging the gap between functional and structural assessment of mastitis. Previous studies have described a wide spectrum of ultrasonographic alterations of the mammary parenchyma in cows affected by mastitis. The normal udder appears as a homogeneous hypoechoic tissue containing anechoic structures corresponding to blood vessels, milk alveoli, and ducts, while the gland cistern is visualized as a large anechoic area containing few hypoechoic dots corresponding to milk [22]. In contrast, inflammatory processes are associated with heterogeneous echotexture, irregular hyperechoic or hypoechoic areas, and structural disruption of the normal parenchymal architecture [23,24,25,26]. Specific sonographic patterns have been reported, such as abscesses (round, well-defined hypoechoic cavities with echogenic walls), fibrotic or gas-filled regions (hyperechoic zones), and hematomas (anechoic to hypoechoic fluid-filled septal spaces) [26]. Furthermore, the presence of non-homogeneous parenchyma has been associated with the likelihood of recovery of milk production following clinical mastitis [24], highlighting the prognostic value of ultrasonographic evaluation. From a practical perspective, these findings may support more accurate prognostic assessments and guide treatment strategies, thereby contributing to a more rational use of antimicrobials [27].
Previous studies have explored teat cistern width and glandular cistern diameter in healthy cows and cows affected by mastitis [7,22,28,29]. Similar investigations have been conducted in small ruminants [30,31,32,33,34]. However, studies specifically addressing the glandular cistern area during both clinical and subclinical mastitis are limited, and the available evidence is often inconsistent, with a lack of consensus on the magnitude and clinical significance of the observed changes.
The limited information available on glandular cistern ultrasonography, together with the reduction in milk production associated with both subclinical and clinical mastitis, prompted the authors of this study to focus on the glandular cistern area during mastitis. The primary aim was therefore to evaluate potential modifications of the glandular cistern area in clinical and subclinical mastitis, and to monitor their evolution over time and during follow-up examinations. Furthermore, the reproducibility of ultrasonographic measurements of the glandular cistern has rarely investigated, and little is known about inter- and intra-operator variability or the reliability of automated image analysis. For this reason, the secondary aim was to assess the level of agreement between operators, intra-operator repeatability, and the accuracy of automated measurements performed using ImageJ.

2. Materials and Methods

2.1. Study Design

A longitudinal study was conducted in Italian Holstein lactating cows from the herd of the Centro di Ricerche Agro-ambientali “E. Avanzi”, University of Pisa. The study was performed in accordance with STROBE guidelines. Ethical approval was granted by the Institutional Animal Care and Use Committee of the University of Pisa (Protocol N: 19/2023, dated 19 April 2023).
Sample size calculation was performed using G*Power 3.1 (Heinrich Heine University, Düsseldorf, Germany). The analysis was based on a repeated measures ANOVA (within–between interaction), assuming an effect size of 0.5, an alpha error probability of 0.05, a power of 0.80, three groups (contralateral healthy, subclinical mastitis, and clinical mastitis quarters), and three repeated measurements (T0, T1, T5). The calculation indicated that a minimum of 12 subjects would be required to achieve adequate power. In the present study, 42 cows (84 quarters) were enrolled, as more animals were available during the study period and the aim was to ensure enough quarters in each group.

2.2. Animals and Management

All enrolled cows were maintained under uniform feeding and housing conditions. Animals were housed in a free-stall barn with permanent straw bedding, fully replaced every 3–4 days, with clean straw added daily. Fresh water was available ad libitum, and feed was distributed twice daily as a total mixed ration. Milking was performed twice daily in a Herringbone parlor. Routine udder health monitoring was carried out weekly by a veterinarian of University of Pisa.

2.3. Inclusion and Exclusion Criteria

Enrollment took place during the weekly veterinary herd checks. Clinical examinations were performed by the veterinarian ready before milking procedures (afternoon), while udder and teats were inspected and palpated for any signs of inflammation once cow entered the milking parlor [34,35]. Before milking, California Mastitis Test (CMT) and somatic cell count (SCC) (DCC DeLaval®, DeLaval s.p.a., Milan, Italy) evaluation were carried out at quarter level. Cows were eligible for inclusion if they met the following criteria: (1) Clinical mastitis (CM): cows with SCC > 200,000 cells/mL, together with visible changes in the udder (swelling, hardness, warmth, pain) and/or alterations in milk appearance (clots, flakes, watery secretion); (2) Subclinical mastitis (SCM): cows with SCC > 200,000 cells/mL, in the absence of clinical signs or visible milk alterations; (3) For each case of CM or SCM, the contralateral quarter of the same cow was required to remain healthy throughout the study period, defined as SCC values below the thresholds indicated above and no clinical or milk alterations. These contralateral healthy quarters constituted the control group.
In addition, a single quarter milk was collected and stored at 4 °C until the bacteriological analysis. For bacteriological examination, 10 µL of milk from each quarter were streaked onto 5% defibrinated sheep blood agar plates. The cultures were incubated aerobically at 37 °C for 24 h, and bacterial growth was evaluated according to colony morphology, hemolytic characteristics, and Gram staining. Representative colonies were subcultured into fresh blood agar to obtain pure isolates. Gram-positive cocci were further differentiated based on catalase and coagulase reactions, whereas Gram-negative bacteria were characterized using colony appearance, Gram staining, oxidase activity, and biochemical reactions on MacConkey agar. Samples yielding growth of three or more distinct bacterial species were classified as contaminated and excluded from pathogen identification.
Cows showing toxic mastitis, i.e., hyperacute cases associated with systemic illness (depression, anorexia, pyrexia, muscle tremors, mucosal congestion) or cows that developed mastitis in the contralateral quarter during follow-up were excluded.
Enrolled cows were monitored for five days and during this period each cows underwent three scheduled ultrasound sessions: T0: day of mastitis diagnosis (afternoon milking). T1: 24 h after diagnosis (afternoon milking) and T5: 5 days after diagnosis (afternoon milking). This time frame was chosen to monitor acute changes during mastitis.

2.4. Ultrasonographic Technique

Transcutaneous ultrasonographic examinations were performed using a portable unit (Mindray DP30Vet, Mindray Bio-Medical Electronics Co., Ltd., Shenzhen, China) equipped with a convex probe operating at 3.5–5 MHz. Images were acquired in B-mode with a scanning depth of 20 cm, gain set at 50 and focus was positioned at approximately 10 cm. The udder was not clipped, and a thin layer of acoustic gel was applied to the skin surface to ensure adequate probe contact. Light pressure was applied during scanning to avoid image distortion. All examinations were carried out during the afternoon, before milking, and performed by the same operator (CO) to avoid variability. The probe was positioned cranially to the teat base. A longitudinal image of the glandular cistern was first obtained by aligning the probe parallel to the teat axis, followed by a transverse image after rotating the probe 90°. Both the affected quarter and the contralateral healthy quarter were examined in each animal. After scanning, residual gel was carefully removed using disposable paper towels. Images were saved on the ultrasound unit, transferred to a USB drive, and archived for subsequent analysis.

2.5. Determination of Glandular Cistern Area

All ultrasound images were then transferred to a computer and processed using ImageJ software v. 1.53 (National Institutes of Health, Bethesda, MD, USA). The analysis of the glandular cistern area was performed in two ways: (1) Manual measurement: the cistern contour was delineated freehand along the echogenic border, contrasting with the anechoic lumen. The software then calculated the enclosed area, expressed in cm2. (2) Automatic measurement: each image was converted to black and white, and the cisternal area was automatically detected using the threshold segmentation function of the software. Image calibration was set so that 24 pixels corresponded to 1 cm, according to the manufacturer’s instructions.
Before evaluation, all images were anonymized to blind the readers to the identity of the animal, and the time point of acquisition. Two independent operators carried out the measurements: Operator A: an experienced veterinarian, diplomate of the European College of Bovine Health Management; Operator B: a veterinarian in the first year of a PhD program. Each operator analyzed the full set of images twice, with a one-month interval between sessions. This approach allowed assessment of intra-operator repeatability. Inter-operator agreement was evaluated by comparing the first reading session of both operators. For each session, both manual and automatic measurements of the cistern area were performed.

2.6. Statistical Analysis

All statistical analyses were performed using SPSS (IBM SPSS Statistics, version 29 IBM Corp., Armonk, NY, USA). Because data were not normally distributed, descriptive statistics were expressed as median, 25th and 75th percentile.
The effect of pathology (healthy, subclinical, clinical) and time (T0, T1, T5) on glandular cistern area was assessed using a two-way mixed ANOVA, with pathology as the between-subjects factor and time as the within-subjects factor. Assumptions of the ANOVA model were formally checked. Mauchly’s test of sphericity was applied to the within-subject factor (time); in case of violation, Greenhouse–Geisser corrected values were considered. Levene’s test was used to assess the equality of error variances across groups. Although the distributions deviated from normality, the two-way mixed ANOVA was retained because it allows simultaneous assessment of time and pathology effects and their interaction, which cannot be addressed by non-parametric alternatives. Moreover, ANOVA models have been demonstrated to be robust to moderate deviations from normality, particularly when group sizes are not extremely different, as in the present dataset [36]. Post-hoc pairwise comparisons were adjusted with Bonferroni correction. Statistical significance was set at a p-value of <0.05.
Intra- and inter-operator reliability of manual and ImageJ-guided acquisitions was evaluated through Intraclass Correlation Coefficients (ICCs). Agreement between operators was classified as poor with ICC values ≤ 0.5, moderate with ICC values between 0.51 and 0.7, good with ICC values between 0.71 and 0.9, and excellent with ICC values > 0.9 [37].

3. Results

All enrolled animals were Italian Friesian dairy cows, ranging from first to seventh lactation, with a median age of 6 years and a median body condition score (BCS) of 3. During the study period, 61 cows were identified with at least one episode of mastitis. Of these, 19/61 cows (30%) were excluded: 14/61 (22%) due to other concomitant diseases (4 lameness, 6 metritis, 4 ketosis) and 5/61 (8%) because mastitis developed in the contralateral control quarter. The final dataset included 42 cows, with 26 subclinical mastitic quarters and 16 clinical mastitic quarters, together with 42 contralateral healthy quarters used as controls. The median milk yield of the included cows was 31 kg (25th p = 14 kg; 75th p = 39 kg). When stratified by health status, milk yield was highest in contralateral healthy quarters (median 33 kg; 25th p = 18 kg; 75th p = 39 kg), slightly reduced in SCM quarters (median 30 kg; 25th p = 16 kg; 75th p = 36 kg), and lowest in CM quarters (median 26 kg; 25th p = 14 kg; 75th p = 32 kg). Cows with SCM were days in milk (DIM) 136.1 ± 103.4 days, while CM cows were 103.9 ± 79.7 DIM. The median number of lactations was 2.5 (25th p = 1; 75th p = 5) in SCM cows and 3.0 (25th p = 2,25; 75th p = 4) in CM cows. The median somatic cell count (SCC) was 104,000 ± 43,000 cells/mL in contralateral healthy quarters, 766,500 ± 443,358 cells/mL in SCM, and 1,627,500 ± 764,794 cells/mL in CM. Bacteriological examination of milk samples confirmed the absence of bacterial growth in all contralateral healthy quarters. Among SCM cases, the predominant isolates were non-aureus staphylococci and Mammaliicoccus spp. (n = 14; 53.8%), followed by Streptococcus dysgalactiae (n = 8; 30.8%), Enterococcus spp. (n = 2; 7.7%), and Aerococcus viridans (n = 2; 7.7%). In quarters with CM, Escherichia coli was the most frequently identified pathogen (n = 9; 56.3%), followed by Streptococcus uberis (n = 6; 37.5%) and Pseudomonas spp. (n = 1; 6.2%).
Ultrasonographic examination of the glandular cistern proved to be feasible under field conditions and was easily performed in all animals. Adequate visualization of the glandular cistern was consistently achieved without the need for shaving, using only ultrasound gel applied directly to the skin. The cistern could be clearly visualized in all quarters at each time point (T0, T1, T5) (Figure 1). The average time required to perform transcutaneous ultrasonography of the glandular cistern was 57 ± 29 s for the longitudinal projection and 46 ± 32 s for the transverse projection. The maximum duration recorded for a complete ultrasonographic examination was 185 s.
Figure 1. Ultrasonographic images of bovine glandular cisterns. (a,b) Healthy quarters in transverse (a) and longitudinal (b) sections. (c,d) Subclinical mastitis (SCM) quarters in transverse (c) and longitudinal (d) sections. (e,f) Clinical mastitis (CM) quarters in transverse (e) and longitudinal (f) sections.
At baseline (T0), the transversal section showed larger cisternal areas in contralateral healthy quarters compared to subclinical and clinical cases. Specifically, the median transversal area was 36.8 mm2 (25.5–44.8) in contralateral healthy quarters, 31.2 mm2 (19.5–38.3) in subclinical cases, and 29.3 mm2 (22.5–42.0) in clinical mastitis. A similar pattern was observed in the longitudinal section, where median cistern areas were 36.6 mm2 (26.8–46.1) in contralateral healthy quarters, 31.6 mm2 (21.4–38.2) in subclinical quarters, and 30.2 mm2 (22.0–39.7) in clinical mastitis. The results of the glandular cistern area across all time points and groups are presented in Table 1.
Table 1. Descriptive statistics (median (IQR)) of glandular cisternal area measurements (mm2) according to clinical group, time, and ultrasonographic section. Different superscript letters (a, b) within the same column indicate significant differences among groups according to post-hoc pairwise comparisons (p < 0.05).
The mixed ANOVA revealed a significant interaction between time and groups both in transversal and longitudinal section (p = 0.043 and 0.034, respectively), indicating that temporal trends differed among healthy, subclinical, and clinical quarters. The between-subjects main effect of group was also significant both in transversal and longitudinal section (p = 0.049 and 0.042, respectively). Post-hoc analysis showed a significant difference between contralateral healthy quarters and those affected by clinical mastitis (transversal section p = 0.034; longitudinal section p = 0.038), while no significant differences were observed between contralateral healthy and subclinical quarters (transversal section p = 0.142; longitudinal section p = 0.132) or between subclinical and clinical quarters (transversal section p = 0.092; longitudinal section p = 0.078).
The agreement in both manual and ImageJ-guided acquisitions showed excellent reliability. Across all strata of pathology (healthy, subclinical, clinical) and time (T0–T5), ICCs were consistently ≥0.95 in the transversal section and ≥0.96 in the longitudinal section.
Both manual and ImageJ-guided acquisitions showed excellent intra-operator reliability. For Operator 1, manual vs. automatic agreement was high (T: 0.959–0.989; L: 0.970–0.991), manual test–retest was among the strongest (T: 0.990–0.996; L: 0.992–0.997), and automatic test–retest reached near-perfect values (T: 0.991–0.999; L: 0.984–0.995). For Operator 2, intra-operator agreement was similarly excellent across methods: manual vs. automatic T 0.981–0.995; L ≥ 0.98; manual test–retest T 0.988–0.997; L ≥ 0.96; automatic test–retest T 0.988–0.995; L 0.981–0.996).
Also inter-operator agreement showed excellent reliability. Manual measurements showed excellent agreement in the transversal section (ICC 0.976–0.999) and very good to excellent agreement in the longitudinal section (0.892–0.998). Automatic acquisitions yielded uniformly excellent inter-operator reliability (T: 0.978–0.994; L: 0.970–0.993). The details of intra- and inter-operator agreement at T0 are reported in Table 2 and Table 3. Complete agreement data for all time points are provided in Tables S1 and S2 of the Supplementary Materials.
Table 2. Intraclass correlation coefficients (ICCs and 95% CI) for intra-operator comparisons (manual and automatic acquisitions) of glandular cistern area at T0, in both transversal (T) and longitudinal (L) sections, stratified by pathology status. Healthy quarters (n = 42), subclinical mastitis (n = 26), and clinical mastitis (n = 16).
Table 3. Intraclass correlation coefficients (ICCs and 95% CI) for inter-operator comparisons (manual and automatic acquisitions) of glandular cistern area at T0, in both transversal (T) and longitudinal (L) sections, stratified by pathology status. Healthy quarters (n = 42), subclinical mastitis (n = 26), and clinical mastitis (n = 16).

4. Discussion

The present study evaluated the use of ultrasonographic assessment of the glandular cistern area during clinical and subclinical mastitis. The first finding was that ultrasound of the glandular cistern is a feasible and reliable method for monitoring structural changes associated with mastitis under field conditions. All images were consistently obtainable without clipping, and visualization of the cisternal lumen was adequate at all time points. In literature, ultrasound of the udder in dairy cows has also been reported as easy to perform [7,38,39]. In recent years, the number of studies employing udder ultrasonography has increased, not only for surgical diseases such as teat obstruction or milk stones [14,40], but also for the evaluation of ultrasound findings during mastitis [27,41].
Our results suggested that cisternal areas were significantly smaller in cows with clinical mastitis compared with contralateral healthy quarters, whereas differences between contralateral healthy and subclinical quarters did not reach statistical significance. This indicates that ultrasonographic alterations become more evident as disease severity increases. The reduction in cisternal size may be explained by inflammatory processes leading to edema, exudation, and tissue thickening, ultimately decreasing the lumen available for milk storage, in association with reduced milk production [42,43]. Previous studies have described echogenic changes in mastitic glands, including increased pa renchymal echogenicity and irregular lumen borders; our results add quantitative evidence on cisternal area reduction as an additional marker of disease [9,15].
In subclinical mastitis, despite elevated SCC values and evidence of inflammation, cisternal measurements showed only modest deviations from contralateral healthy quarters. This observation aligns with the subclinical nature of the disease, in which functional impairment may precede or exceed detectable structural changes [9,15]. Longer follow-up and larger cohorts could help clarify whether cisternal changes evolve over time or in relation to specific etiologies. Beyond structural assessment, Doppler ultrasonography offers the possibility to evaluate hemodynamic and lymphatic alterations associated with mastitis. Studies have demonstrated changes in blood flow parameters and regional lymph node perfusion in affected cows [44,45]. Integrating these findings with B-mode measurements could improve understanding of disease dynamics. Furthermore, recent research has explored the application of machine-learning and deep-learning models to ultrasonographic and sensor data for automated mastitis detection [46,47]. These advances suggest a potential for combining Doppler and artificial-intelligence-based analyses to enhance diagnostic accuracy and support precision dairy management.
The interaction between time and pathology observed in our study underscores that temporal dynamics differ according to disease status. In clinical mastitis, a marked decrease in cisternal area at T1 was followed by partial recovery at T5, suggesting that ultrasonography may be useful to monitor disease progression and response to treatment. Conversely, contralateral healthy and subclinical quarters showed relatively stable measurements over time. These findings are consistent with udder damage caused by clinical mastitis and with the effects described in previous literature [48,49]. In fact, clinical mastitis has a direct effect on the udder parenchyma, influencing milk production throughout the entire lactation in dairy cows [50,51]. This mechanism could explain the differences in cistern area observed during the study period.
A notable finding of the present study was the excellent reproducibility of ultrasonographic measurements of the glandular cistern area. Both manual and automated (ImageJ-guided) analyses achieved ICC values consistently above 0.95 across operators, reading sessions, and pathology groups. Such values fall within the range generally classified as “excellent” according to established reliability criteria [37]. To the authors’ knowledge, no previous study has assessed agreement for the glandular cistern area in dairy cows. The very high intra-operator repeatability observed suggests that both manual tracing and automated segmentation can be reliably reproduced by the same examiner over time. This is crucial in longitudinal designs, where repeated measures are required to monitor the evolution of mastitis within the same cow or herd. At the same time, the excellent inter-operator agreement highlights that ultrasonographic evaluation of the cistern can be standardized across different examiners with minimal variability, making it suitable for multicentric studies and for routine herd monitoring performed by different veterinarians. The high level of both inter- and intra-operator agreement observed in this study supports the applicability of this ultrasonographic approach under field conditions. In contrast, previous investigations on the use of ultrasonography for bovine respiratory disease have reported disappointing reproducibility, largely attributed to limited operator expertise [52]. Interestingly, when applied to the teat, such constraints appear less critical, provided that the examiner has received at least basic theoretical training, which seems sufficient to ensure reliable results.
Automated image analysis deserves particular attention. The use of ImageJ to segment the cistern lumen and compute its area further minimized variability, producing near-perfect ICC values in both transversal and longitudinal sections. This suggests that, once appropriate thresholds and calibration are established, automated methods may outperform manual tracings in terms of consistency. In the literature, automated image analysis in both human and veterinary medicine has been reported as a method to improve agreement and reproducibility of results [9,53,54]. From a practical perspective, it may reduce operator workload by shortening the time needed for measurements, a relevant advantage when large datasets are analyzed.
Nevertheless, some limitations should be acknowledged. The study was restricted to a single herd, which may limit the generalizability of the findings, and the relatively small number of clinical mastitis cases may have reduced statistical power for subgroup comparisons. The unbalanced distribution among groups (CM n = 16, SCM n = 26, healthy n = 42) should also be recognized as a limitation. Although the overall number of enrolled quarters exceeded the minimum sample size required by the power analysis, the smaller CM group may have increased variability and reduced the precision of between-group estimates, particularly for interaction effects.
Bacteriological culture results were available but were not included in the statistical analysis, as the study was primarily designed to evaluate ultrasonographic differences according to clinical classification rather than etiological agents. Consequently, pathogen-specific effects could not be assessed. Different microorganisms may induce variable inflammatory patterns and involve distinct mammary compartments, potentially influencing cisternal ultrasonographic appearance. Future studies should therefore investigate whether specific bacterial species are associated with characteristic ultrasonographic findings.
Moreover, although automated image analysis yielded promising results, some caution should be exercised when interpreting these data. Automated segmentation algorithms are influenced by image quality, probe positioning, and artifacts such as debris or acoustic shadowing, which may affect reproducibility under field conditions.
Finally, cows showing systemic involvement or toxic mastitis were excluded to avoid potential confounding factors related to fever, dehydration, or severe inflammation, which could independently alter milk yield and cisternal dimensions. This restriction allowed a clearer evaluation of the method’s reliability under comparable conditions but limits the applicability of the results to mild or moderate mastitis.
Future research should include animals with systemic or severe forms of mastitis to define diagnostic cut-offs and to evaluate the sensitivity and specificity of ultrasonographic cisternal measurements. Further studies are also warranted to assess whether cisternal alterations vary according to the infectious agent, disease chronicity, or parity, and to validate the robustness of automated analyses under variable field conditions. Finally, integrating ultrasonography with functional parameters such as SCC, milk yield, and biochemical indicators could enhance diagnostic accuracy and provide a more comprehensive understanding of mastitis progression.

5. Conclusions

In conclusion, this study confirms that ultrasonographic measurement of the glandular cistern area is a practical, reproducible, and informative tool for mastitis evaluation. While clinical mastitis is associated with significant cisternal reduction, changes in subclinical cases remain limited. The robustness of both manual and automated measurements supports the adoption of this technique for longitudinal monitoring and research purposes, with promising implications for precision dairy farming.

Supplementary Materials

The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/dairy6060068/s1, Table S1. Intraclass correlation coefficients (ICCs and 95% CI) for transversal section measurements of glandular cistern area, stratified by pathology status (healthy, subclinical, clinical quarters) and time (T0–T2). Agreement was evaluated across eight comparisons: intra-operator agreement for manual vs. automatic acquisition, manual test–retest, automatic test–retest; inter-operator agreement for manual vs. manual and automatic vs. automatic measurements. Healthy quarters (n = 42), subclinical mastitis (n = 26), and clinical mastitis (n = 16). Table S2. Intraclass correlation coefficients (ICCs, average measures, 95% CI) for longitudinal section measurements of glandular cistern area, stratified by pathology status (healthy, subclinical, clinical quarters) and time (T0–T2). Agreement was evaluated across eight comparisons: intra-operator agreement for manual vs automatic acquisition, manual test–retest, automatic test–retest; inter-operator agreement for manual vs. manual and automatic vs. automatic measurements. Healthy quarters (n = 42), subclinical mastitis (n = 26), and clinical mastitis (n = 16).

Author Contributions

Conceptualization, G.S. and F.B.; methodology, G.S.; software, G.S.; validation, G.S., C.O. and F.B.; formal analysis, G.S.; investigation, G.S., C.O., A.C. and F.B.; data curation, G.S.; writing—original draft preparation, M.C. and G.S.; writing—review and editing, F.B.; visualization, G.A.; supervision, F.B. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Institutional Review Board Statement

Ethical approval was granted by the Institutional Animal Care and Use Committee of the University of Pisa (Protocol N: 19/2023, dated 19 April 2023).

Data Availability Statement

The original contributions presented in this study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding author.

Conflicts of Interest

The authors declare no conflicts of interest.

Abbreviations

The following abbreviations are used in this manuscript:
ANOVAAnalysis of variance
BCSBody condition score
CMTCalifornia Mastitis Test
CMClinical mastitis
CI Confidence interval
ICC(s) Intraclass correlation coefficient(s)
IQR Interquartile range
SCM Subclinical mastitis
SCC Somatic cell count

References

  1. Fricke, P.M. Scanning the future—Ultrasonography as a reproductive management tool for dairy cattle. J. Dairy Sci. 2002, 85, 1918–1926. [Google Scholar] [CrossRef]
  2. Ollivett, T.L.; Buczinski, S. On-farm use of ultrasonography for bovine respiratory disease. Vet. Clin. N. Am. Food Anim. Pract. 2016, 32, 19–35. [Google Scholar] [CrossRef] [PubMed]
  3. Braun, U.; Reif, C.; Nuss, K.; Hilbe, M.; Gerspach, C. Clinical, laboratory and ultrasonographic findings in 87 cows with type-4 abomasal ulcer. BMC Vet. Res. 2019, 15, 100. [Google Scholar] [CrossRef]
  4. Fiore, E.; Faillace, V.; Morgante, M.; Armato, L.; Gianesella, M. A retrospective study on transabdominal ultrasound measurements of the rumen wall thickness to evaluate chronic rumen acidosis in beef cattle. BMC Vet. Res. 2020, 16, 337. [Google Scholar] [CrossRef]
  5. Pravettoni, D.; Buczinski, S.; Sala, G.; Ferrulli, V.; Bianchi, F.; Boccardo, A. Diagnostic accuracy of focused lung ultrasonography as a rapid method for the diagnosis of respiratory disease in dairy calves. J. Dairy Sci. 2021, 104, 4929–4935. [Google Scholar] [CrossRef] [PubMed]
  6. Szencziová, I.; Strapák, P. Ultrasonography of the udder and teat in cattle: Perspective measuring technique. Slovak J. Anim. Sci. 2012, 45, 96–104. [Google Scholar]
  7. Bonelli, F.; Orsetti, C.; Turini, L.; Meucci, V.; Pierattini, A.; Sgorbini, M.; Citi, S. Mammary cistern size during the dry period in healthy dairy cows: A preliminary study for an ultrasonographic evaluation. Animals 2020, 10, 2082. [Google Scholar] [CrossRef] [PubMed]
  8. Vang, A.L.; Frizzarini, W.S.; Webster, H.H.; Cunha, T.O.; Nelson, K.M.; Dorea, J.R.; Hernandez, L.L. Ultrasound-assisted minimally invasive technique for tissue collection from bovine mammary glands. Vet. Med. Sci. 2024, 10, e70027. [Google Scholar] [CrossRef]
  9. Tommasoni, C.; Lisuzzo, A.; Fiore, E.; Taio, G.; Cecchini, F.; Gottardo, A.; Gianesella, M. Ultrasound evaluation of mammary gland cistern in quarters affected by mastitis: New diagnostic grading. Large Anim. Rev. 2025, 31, 167–172. [Google Scholar]
  10. Nocera, I.; Sala, G.; Orsetti, C.; Rodriguez-Lopez, J.M.; Sgorbini, M.; Bonelli, F. Evaluation of intra- and inter-rater agreement on ultrasound measurements of dairy cow teats by operators with different experience levels. Vet. Res. Commun. 2025, 49, 9. [Google Scholar] [CrossRef]
  11. Olechnowicz, J.; Jaśkowski, J.M. Ultrasound examination of mammary glands in ruminants. Med. Weter. 2009, 65, 147–150. [Google Scholar]
  12. Barbagianni, M.S.; Mavrogianni, V.S.; Vasileiou, N.G.C.; Fthenakis, G.C.; Petridis, I.G. Ultrasonographic examination of the udder in sheep. Small Rumin. Res. 2017, 152, 86–99. [Google Scholar] [CrossRef]
  13. Kotb, E.E.Z.; Fadel, M.; Abd El-Fattah, O.A.; Azab, A.M.S.; Leil, A.Z.A. Ultrasonography, histopathological, udder alterations and bacteriological investigations for diagnosis of mastitic goats. J. Appl. Vet. Sci. 2020, 5, 77–86. [Google Scholar] [CrossRef]
  14. Franz, S.; Floek, M.; Hofmann-Parisot, M. Ultrasonography of the bovine udder and teat. Vet. Clin. N. Am. Food Anim. Pract. 2009, 25, 669–685. [Google Scholar] [CrossRef]
  15. Javadi, T.; Acorda, J.A. Ultrasound features and echo mean values of udder and teat in dairy cows with mastitis. Philipp. J. Vet. Anim. Sci. 2011, 37, 149–156. [Google Scholar]
  16. Bae, H.; Jeong, C.H.; Cheng, W.N.; Hong, K.; Seo, H.G.; Han, S.G. Oxidative stress-induced inflammatory responses and effects of N-acetylcysteine in bovine mammary alveolar cells. J. Dairy Res. 2017, 84, 418–425. [Google Scholar] [CrossRef] [PubMed]
  17. Hogeveen, H.; Huijps, K.; Lam, T.J.G.M. Economic aspects of mastitis: New developments. N. Z. Vet. J. 2011, 59, 16–23. [Google Scholar] [CrossRef]
  18. Lakshmi, K.; Jayavardhanan, K. Screening of milk samples for sub-clinical and clinical mastitis by using CMT and SCC. J. Med. Sci. Clin. Res. 2016, 4, 10853–10855. [Google Scholar] [CrossRef]
  19. Harmon, R.J. Physiology of mastitis and factors affecting somatic cell counts. J. Dairy Sci. 1994, 77, 2103–2112. [Google Scholar] [CrossRef]
  20. Cobirka, M.; Tancin, V.; Slama, P. Epidemiology and classification of mastitis. Animals 2020, 10, 2212. [Google Scholar] [CrossRef]
  21. Schukken, Y.; Wilson, D.; Welcome, F.; Garrison-Tikofsky, L.; Gonzalez, R. Monitoring udder health and milk quality using somatic cell counts. Vet. Res. 2003, 34, 579–596. [Google Scholar] [CrossRef]
  22. Mourya, A.; Shukla, P.C.; Gupta, D.K.; Sharma, R.K.; Nayak, A.; Tiwari, A.; Singh, B.; Singh, A.P.; Sahi, A.; Jain, A. Ultrasonographic alteration in subclinical mastitis in cows. J. Entomol. Zool. Stud. 2020, 8, 2058–2063. [Google Scholar]
  23. Fasulkov, I.R. Ultrasonography of the Mammary Gland in Ruminants: A Review. Bulg. J. Vet. Med. 2012, 15, 1–12. [Google Scholar]
  24. Suzuki, N.; Kurose, T.; Kaneko, S.; Haraguchi, A.; Isobe, N. Outcome Prediction from the First Examination in Clinical Mastitis Using Ultrasonography in Dairy Cows. Anim. Sci. J. 2020, 91, e13452. [Google Scholar] [CrossRef]
  25. Sunder, H.A.; Gupta, D.; Singh, R.; Singh, S.; Randhawa, C. Ultrasonographic Changes in Teat and Supramammary Lymph Nodes in Dairy Cows Affected with Clinical Mastitis. Haryana Vet. 2022, 61, 68–71. [Google Scholar]
  26. Flöck, M.; Winter, P. Diagnostic Ultrasonography in Cattle with Diseases of the Mammary Gland. Vet. J. 2006, 171, 314–321. [Google Scholar] [CrossRef]
  27. Tommasoni, C.; Fiore, E.; Lisuzzo, A.; Gianesella, M. Mastitis in dairy cattle: On-farm diagnostics and future perspectives. Animals 2023, 13, 2538. [Google Scholar] [CrossRef] [PubMed]
  28. Bruckmaier, R.M.; Rothenanger, E.; Blum, J.W. Measurement of mammary gland cistern size and determination of the cisternal milk fraction in dairy cows. Milchwissenschaft 1994, 49, 543–546. [Google Scholar]
  29. Ayadi, M.; Caja, G.; Such, X.; Knight, C.H. Use of ultrasonography to estimate cistern size and milk storage at different milking intervals in the udder of dairy cows. J. Dairy Res. 2003, 70, 1–7. [Google Scholar] [CrossRef]
  30. Nudda, A.; Pulina, G.; ValleBella, R.; Bencini, R.; Enne, G. Ultrasound technique for measuring mammary cistern size of dairy ewes. J. Dairy Res. 2000, 67, 101–106. [Google Scholar] [CrossRef] [PubMed]
  31. Milerski, M.; Margetín, M.; Capistrak, A.; Apolen, D.; Spanik, J.; Oravcová, M. Relationships between external and internal udder measurements and the linear scores for udder morphology traits in dairy sheep. Czech J. Anim. Sci. 2006, 51, 383–390. [Google Scholar] [CrossRef]
  32. Rovai, M.; Caja, G.; Such, X. Evaluation of udder cisterns and effects on milk yield of dairy ewes. J. Dairy Sci. 2008, 91, 4622–4629. [Google Scholar] [CrossRef] [PubMed]
  33. Makovický, P.; Margetín, M.; Milerski, M. Estimation of udder cistern size in dairy ewes by ultrasonography. Mljekarstvo 2015, 65, 210–218. [Google Scholar] [CrossRef]
  34. Andrews, A.H.; Blowey, R.W.; Boyd, H. Bovine Medicine: Diseases and Husbandry of Cattle; John Wiley & Sons: Oxford, UK, 2007; p. 1232. [Google Scholar]
  35. Blowey, R.; Edmonson, P. Mastitis Control in Dairy Herds, 2nd ed.; CABI: Wallingford, UK, 2010; pp. 1–208. [Google Scholar]
  36. Glass, G.V.; Peckham, P.D.; Sanders, J.R. Consequences of failure to meet assumptions underlying the fixed effects analyses of variance and covariance. Rev. Educ. Res. 1972, 42, 237–288. [Google Scholar] [CrossRef]
  37. Koo, T.K.; Li, M.Y. A guideline of selecting and reporting intraclass correlation coefficients for reliability research. J. Chiropr. Med. 2016, 15, 155–163. [Google Scholar] [CrossRef]
  38. Themistokleous, K.S.; Papadopoulos, I.; Panousis, N.; Zdragas, A.; Arsenos, G.; Kiossis, E. Udder ultrasonography of dairy cows: Investigating the relationship between echotexture, blood flow, somatic cell count and milk yield during dry period and lactation. Animals 2023, 13, 1779. [Google Scholar] [CrossRef]
  39. Schwarz, T.; Scheeres, N.; Małopolska, M.M.; Murawski, M.; Agustin, T.D.; Ahmadi, B.; Bartlewski, P.M. Associations between mammary gland echotexture and milk composition in cows. Animals 2020, 10, 2005. [Google Scholar] [CrossRef] [PubMed]
  40. Ragab, G.H.; Seif, M.M.; Qutp, M.M. Ultrasonography of the mammary gland in ruminants. J. Vet. Med. Res. 2016, 23, 125–132. [Google Scholar]
  41. Abdullah, O.M.; Aslam, S.; Khan, M.A.; Mushtaq, H.; Hassan, M.; Ijaz, M. Healthy, sub-clinical, and clinical mastitis in Holstein–Friesian cattle: A comparative echotextural and electrical conductivity study. S. Afr. J. Anim. Sci. 2023, 53, 221–230. [Google Scholar]
  42. Günther, J.; Petzl, W.; Bauer, I.; Ponsuksili, S.; Zerbe, H.; Schuberth, H.J.; Seyfert, H.M. Differentiating Staphylococcus aureus from Escherichia coli mastitis: S. aureus triggers unbalanced immune-dampening and host cell invasion immediately after udder infection. Sci. Rep. 2017, 7, 4811. [Google Scholar] [CrossRef]
  43. Ameni, G.; Bayissa, B.; Zewude, A.; Degefa, B.A.; Mohteshamuddin, K.; Kalaiah, G.; Tibbo, M. Retrospective study on bovine clinical mastitis and associated milk loss during the month of its peak occurrence at the National Dairy Farm in the Emirate of Abu Dhabi, United Arab Emirates. Front. Vet. Sci. 2022, 9, 1070051. [Google Scholar] [CrossRef]
  44. Rişvanli, A.; Dogan, H.; Safak, T.; Kilic, M.A.; Seker, I. The relationship between mastitis and the B-mode, colour Doppler ultrasonography measurements of supramammary lymph nodes in cows. J. Dairy Res. 2019, 86, 315–318. [Google Scholar] [CrossRef] [PubMed]
  45. Rişvanli, A.; Dogan, H.; Safak, T.; Kilic, M.A.; Seker, I. Pulsed-wave Doppler ultrasonographical measurements of supra-mammary lymph nodes for diagnosis of bovine mastitis. Vet. Res. Forum 2022, 13, 431. [Google Scholar] [CrossRef]
  46. Zhang, X.; Li, Y.; Zhang, Y.; Yao, Z.; Zou, W.; Nie, P.; Yang, L. A new method to detect buffalo mastitis using udder ultrasonography based on deep learning network. Animals 2024, 14, 707. [Google Scholar] [CrossRef]
  47. Tian, H.; Zhou, X.; Wang, H.; Xu, C.; Zhao, Z.; Xu, W.; Deng, Z. The prediction of clinical mastitis in dairy cows based on sensor data and machine learning algorithms. Animals 2024, 14, 315. [Google Scholar] [CrossRef] [PubMed]
  48. Goulart, D.B.; Mellata, M. Escherichia coli mastitis in dairy cattle: Etiology, diagnosis, and treatment challenges. Front. Microbiol. 2022, 13, 928346. [Google Scholar] [CrossRef]
  49. Choudhary, R.K.; Olszanski, L.; McFadden, T.B.; Lalonde, C.; Spitzer, A.; Shangraw, E.M.; Zhao, F.Q. Systemic and local responses of cytokines and tissue histology following intramammary lipopolysaccharide challenge in dairy cows. J. Dairy Sci. 2024, 107, 1299–1310. [Google Scholar] [CrossRef]
  50. Sharun, K.; Dhama, K.; Tiwari, R.; Gugjoo, M.B.; Iqbal Yatoo, M.; Patel, S.K.; Chaicumpa, W. Advances in therapeutic and managemental approaches of bovine mastitis: A comprehensive review. Vet. Q. 2021, 41, 107–136. [Google Scholar] [CrossRef] [PubMed]
  51. Zhao, X.; Lacasse, P. Mammary tissue damage during bovine mastitis: Causes and control. J. Anim. Sci. 2008, 86, 57–65. [Google Scholar] [CrossRef]
  52. Jourquin, S.; Lowie, T.; Bokma, J.; Pardon, B. Accuracy and inter-rater agreement among practitioners using quick thoracic ultrasonography to diagnose calf pneumonia. Vet. Rec. 2024, 194, e3896. [Google Scholar] [CrossRef]
  53. Castro, D.; Baba, L.H.K.; Butterfield, M.; Castro, L. Quantitative analysis of ultrasound-guided sciatic nerve staining: ImageJ software application. Vet. Anaesth. Analg. 2024, 51, 354–356. [Google Scholar] [CrossRef] [PubMed]
  54. Qureshi, A.; Chen, H.; Saeedi, O.; Kaleem, M.A.; Stoleru, G.; Margo, J.; Alexander, J.L. Anterior segment ultrasound biomicroscopy image analysis using ImageJ software: Intra-observer repeatability and inter-observer agreement. Int. Ophthalmol. 2019, 39, 829–837. [Google Scholar] [CrossRef] [PubMed]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Article Metrics

Citations

Article Access Statistics

Journal Statistics
Multiple requests from the same IP address are counted as one view.
Download PDF
Influence of Dietary and Ruminal Factors on Microbial and Non-Microbial Nitrogen Flows to the Small Intestine in Lactating Dairy Cows: A Meta-Analysis
Previous
Bioconversion of Deproteinized Cheese Whey to Metabolites by Understudied Cryptococcus-Related Yeasts: Characterization and Properties of Extracted Polysaccharides
Next