Cows with High SCC Exhibit Poorer Performance and Milk Quality, Regardless of the Season

Abstract

This study aimed to examine the relationship between a high somatic cell count (SCC) in cows and milk quality during the hot season in different breeds. Milk samples from 500 cows in the hot season and 431 in the cold season of 2022 were collected across 39 farms in Santa Catarina, Brazil. The samples were analyzed for SCC, milk composition, and physical attributes. Cows were also evaluated for udder depth, udder clearance, teat-end condition, and leg and udder cleanliness. Based on the SCC levels, cows were categorized as low (≤200,000 cells/mL), medium (>200,000 and ≤615,000), or high (>615,000). Data were analyzed by ANOVA with a statistical model that included the effects of the SCC class, season, days in milk, parity, genetic group, and the interaction of the SCC level and season. The results showed that cows with a high SCC produced less milk with lower component levels but higher chloride content. Milk from the hot season had lower acidity and reduced component levels. The impact of SCC on the physical traits of milk did not vary with season. Furthermore, cows with deeper udders and lower udder clearance were more likely to have a high SCC, regardless of genetics. Both a high SCC and hot temperatures independently compromised milk yield and quality, thereby increasing the risk of culling. Therefore, improving udder conformation and avoiding cows with deep udders may help to reduce SCC levels.

1. Introduction

The high temperature and relative humidity found in subtropical regions hinder the thermoregulation of lactating cows, particularly in the hot season [1,2]. Although we know that the increase in the temperature–humidity index (THI) has been associated with lower milk production [3], this difficulty in achieving thermal balance may also favor heat stress and increase the somatic cell count (SCC) in milk [2,4]. This consequence may be associated with the increased survival of pathogens in warm climate regions, the weaker immune response of cows [5], or reduced feed intake [4]. Moreover, characteristics of the cows such as days in milk (DIM), parity, the condition of the teat-ends, and the cleanliness of the udder can also predispose them to the development of mastitis [6].

Cows with a high SCC reduce milk yield (MY) and alter the milk’s physicochemical quality [2,7,8,9]. In Brazil, the milk’s physical attributes are regulated by Normative Instruction 76 [10], with non-compliance with the parameters leading to financial losses for dairy farms. Evidence suggests that a high SCC impairs milk ethanol stability (MES) [11,12], causes low acidity [9], low lactose content [7,13,14], low protein content [7,13], low fat content [7], and reduces the milk freezing point (FP) [15,16]. Moreover, there is an increase in some ions (Na⁺ and Cl⁻) in the alveolar lumen [17], which increases milk chloride content [18]. Despite this, knowledge about the influence of SCC on milk attributes still appears to be inconclusive, especially regarding the physical attributes of milk.

Heat stress may also influence the milk’s physicochemical attributes; nevertheless, its relationship with the SCC in milk is still not fully understood. Previous studies suggest a lower MY, reduced physicochemical quality of the milk, MES [3], low acidity, and positive reactions in the qualitative chloride test [9] in a high THI. Other studies have investigated the influence of the time of year on milk quality [19,20,21]. Bernabucci et al. [19] identified the lowest concentrations of fat and protein in milk in summer, while the highest concentrations were found in winter, with intermediate levels in spring. Furthermore, they suggested a possible increase in the SCC during summer.

Similarly, there is evidence that the physicochemical attributes are altered during summer [19,20,21]. Bjerg et al. [20] suggest that the higher FP may be influenced by dietary changes, temperature, and increased water intake following dehydration. The literature indicates that MES is impaired during respiratory alkalosis [17], heat stress [11,22], and by the nutritional status of lactating cows [22,23]. Furthermore, the physicochemical quality of milk is influenced by other characteristics of the cow, such as DIM [9,21] and parity [9].

Thus far, it is known that the physicochemical quality of milk can be affected by an elevated SCC and by hot climatic conditions. In subtropical climate regions, climatic factors pose challenges to milk production, especially for cows on pasture. Our objective was to identify whether the performance of dairy cows, as well as the physicochemical attributes of milk, are affected by subclinical mastitis and the hot climatic season.

2. Materials and Methods

The experiment was conducted from January to August 2022 on 39 commercial farms with a pasture-based system in western Santa Catarina, Brazil. It is a subtropical climate region of the humid temperate type with hot summers, according to Köppen [1]. The farms were selected based on the geometric mean of the tank SCC corresponding to the three months prior to the scheduled date of the farm visit. For the farm to be visited, the geometric mean had to be above 500,000 cells/mL to ensure the presence of cows with both high and low SCCs within the same herd. The farms had an average of 22.00 ± 1.27 lactating cows.

Most farms also provided roughage and protein concentrate supplementation. The main roughage supplement was maize silage (Zea mays), although some farms also supplemented cows with haylage. The protein concentrate was typically a commercial formulation, although some farms mixed it themselves. The study was conducted on small farms with an average area of 29 hectares (ranging from 5.8 to 98 hectares) and an average of 22.6 lactating cows (ranging from 7 to 58 cows), producing an average of 16.07 kg/cow/day (ranging from 3 to 45 kg). The herd composition consisted mainly of Holstein (47.43%), Jersey (38.51%), and crossbred cows (14.06%), which are representative of milk production in this region.

In each season, we collected one milk sample from each cow, totaling 500 cows evaluated in the hot season (January, February, and March) and 431 cows evaluated in the cold season (June, July, and August), with the same person collecting milk samples from the available cows at each farm. The following information was also recorded: genetic group, DIM, and parity. During one of the milking sessions, udder depth was measured as the distance (cm) from the base of the udder to the hock line [24]. At the same time, udder clearance was measured by the distance (cm) from the base of the udder to the ground [25]. Leg and udder cleanliness were scored using the following scale: 1 (completely clean), 2 (slightly dirty), 3 (dirty), or 4 (completely dirty) [26]. After the detachment of the milking unit, the condition of the teat-end was scored as 1 (no ring), 2 (smooth or slightly rough ring), 3 (rough ring), or 4 (very rough ring) [27].

Milk samples were collected from individual cows during milking using Waikato Milk Meter (Waikato Milking Systems, Hamilton, New Zealand) to identify the daily MY. In the bucket milking system, sampling was carried out after weighing and homogenizing the milk in the pail. All farms performed two milkings within a 24 h period. After the first milk collection, the samples were kept in plastic flasks at 4 °C until the second collection. The proportional sample from each milking was divided between two 40 mL plastic flasks with lids, one containing a preservative Bronopol^® (2-bromo-2-nitropropane-1,3-diol) and the other without a preservative, and all were kept refrigerated. Flasks with preservative were shipped the day after collection to the Paraná Association of Holstein Cattle Breeders laboratory for milk composition, FP, and SCC determination.

The concentrations of fat, protein, casein, lactose, defatted dry extract (DDE), total solids extract (TSE), and FP were determined by Fourier Transform Infrared Spectroscopy (Bentley Instruments, Chaska, MN, USA). The SCC was determined by flow cytometry using the Somacount 300 equipment (Bentley Instruments, Chaska, MN, USA). The fat-to-protein ratio, casein-to-protein ratio, and protein and fat yield were calculated. The average nergy-corrected milk (ECM, kg/cow/day) was estimated according to Kirchgeßner, as follows [28]: ECM (kg) = MY (kg) × {[0.39 × fat% + 0.24 × protein% + 0.17 × lactose%]/3.17}.

Flasks without conservative were cooled at 4 °C for 12 h and analyzed for MES, pH, acidity, and chloride content. The MES was determined by mixing 2 mL of milk and 2 mL of alcoholic solutions with ethanol concentrations of 68, 70, 72, 74, 76, 78, 80, 82, 84, 86, and 88% v/v in a Petri dish [28]. The results were expressed as the minimum ethanol concentration in the alcoholic solution that induced visually detected milk coagulation [29]. The pH and acidity were determined by potentiometry (Akso, model K39-6322PHC, Petaling Jaya, Malaysia) and titration, respectively [30]. Acidity was expressed as the g of lactic acid per 100 mL of milk. The chloride content (g/L) was determined by homogenizing 10 mL of milk, 125 mL of distilled water, and two drops of nitric acid [31]. The mixture was titrated with silver nitrate (0.08 mol/L) until the lowest conductivity (Akso, model K39-003PHC). At the point of lowest conductivity, the titration was stopped, and the volume of silver nitrate (V, in mL) was recorded. The chloride content (g/L) was determined using the following equation: chloride content = {[0.142 × V] × 2}.

For statistical analysis, we classified the cows into three SCC classes. The first one (low SCC) included 392 cows without subclinical mastitis (≤200,000 cells/mL), and the cows with subclinical mastitis (>200,000 cells/mL) were divided into medium (n = 297) and high SCC (n = 249) groups by a threshold of 615,000 cells/mL. This threshold was set in order to avoid groups with a similar number of cows. All 39 herds had cows in each class of SCC.

The data of traits related to milk yield, physicochemical attributes of milk, udder depth, and mean hyperkeratosis of teat-end were analyzed using ANOVA with the GLM procedure of the SAS statistical package (version 9.3, SAS Institute Inc., Cary, NC, USA). The normality of residuals was tested using the UNIVARIATE procedure, with a significance level of 5%, based on the Kolmogorov–Smirnov test. The statistical model included the effects of the SCC class of the cows (low, medium, and high), the season (hot and cold), the DIM (<100, 100–200, 200–300, >300), the parity (1, 2, 3, and ≥4), the genetic group (Jersey, crossbred dairy cows, Holstein), the dairy farm, and the interaction between the SCC class and season. The means were compared by Tukey’s test, and means were considered significantly different when p < 0.05. The relationship between the SCC classes and hyperkeratosis (teat-end condition with a score of 3 or higher) was analyzed using the chi-square test. Variables that did not meet the assumption of normality were analyzed using the nonparametric Kruskal–Wallis test, implemented through the NPAR1WAY procedure of the SAS statistical package.

In parallel with the univariate statistical analyses, a multivariate analysis was performed using factor analysis to evaluate the relationships among variables related to the cows and milk quality. The analysis was conducted using the FACTOR procedure of the SAS statistical package, with a Promax matrix rotation, and the data were standardized using the STANDARD procedure. Factor loadings ≥ 0.4 were considered significant.

3. Results

Healthy cows (≤200,000 cells/mL) had a mean of 80,500 cells/mL of milk, while those with subclinical mastitis had a mean of 372,910 and 1877,280 cells/mL of milk for medium and high SCC groups, respectively (

Table 1. Means ± mean square error and p-values of milk yield and milk attributes in lactating dairy cows according the SCC class (low, medium, and high) and season (cold and hot).

Variables	SCC Class			Season		p-Value 1
	Low (n = 392)	Medium (n = 297)	High (n = 249)	Cold (n = 431)	Hot (n = 500)	SCC 2	Season	DIM 3	Genetic Group 4	Parity 5	SCC x Season
Milk yield, kg/day	16.05 ± 0.19 a	15.19 ± 0.22 b	14.23 ± 0.24 c	15.84 ± 0.18	14.47 ± 0.17	***	*	***	***	***	*
ECM kg/day 6	18.35 ± 0.21 a	17.46 ± 0.25a	16.30 ± 0.27 b	18.60 ± 0.20	16.14 ± 0.19	***	***	***	*	***	NS
SCC, ×1000 cells/mL	80.50	372.91	1877.28	769.66	784.14	-	-	-	-	-	-
Fat, g/100 g	4.42 ± 0.04	4.42 ± 0.03	4.30 ± 0.03	4.55 ± 0.03	4.15 ± 0.02	NS	***	***	***	NS	*
Protein, g/100 g	3.51 ± 0.01 b	3.59 ± 0.02 a	3.59 ± 0.02 a	3.59 ± 0.01	3.54 ± 0.01	*	NS	***	***	*	NS
Lactose, g/100 g	4.46 ± 0.01 a	4.34 ± 0.01 b	4.15 ± 0.01 c	4.35 ± 0.01	4.29 ± 0.01	***	NS	***	*	***	NS
Casein, g/100 g	2.79 ± 0.01 b	2.87 ± 0.02 a	2.86 ± 0.02 a	2.87 ± 0.01	2.81 ± 0.01	**	NS	***	***	*	NS
Total solids extract, g/100 g	13.32 ± 0.04 a	13.38 ± 0.05 a	13.05 ± 0.06 b	13.52 ± 0.04	12.98 ± 0.04	**	***	***	***	*	NS
Defatted dry extract, g/100 g	8.93 ± 0.02 a	8.93 ± 0.02 a	8.74 ± 0.02 b	8.96 ± 0.02	8.77 ± 0.02	***	**	***	***	***	NS
Fat-to-protein ratio, g/100 g	1.23 ± 0.01 a	1.22 ± 0.01 ab	1.19 ± 0.01 b	1.26 ± 0.01	1.17 ± 0.01	*	**	NS	NS	NS	NS
Casein-to-protein ratio, g/100 g	0.80 ± 0.00 a	0.80 ± 0.00 b	0.80 ± 0.00 b	0.80 ± 0.00	0.80 ± 0.00	*	*	**	NS	NS	NS
Chloride content, g/L	0.85 ± 0.01 c	0.93 ± 0.01 b	1.07 ± 0.01 a	0.91 ± 0.01	0.99 ± 0.01	***	**	NS	**	***	NS
Fat yield, kg/day	0.68 ± 0.01 a	0.65 ± 0.01 a	0.59 ± 0.01 b	0.69 ± 0.01	0.58 ± 0.01	***	***	***	NS	***	*
Protein yield, kg/day	0.55 ± 0.01 a	0.53 ± 0.01 a	0.50 ± 0.01 b	0.56 ± 0.01	0.50 ± 0.01	***	**	***	*	***	NS
Lactose yield, kg/day	0.72 ± 0.01 a	0.66 ± 0.01 b	0.60 ± 0.01 c	0.69 ± 0.01	0.63 ± 0.01	***	*	***	***	***	NS
MES, °GL 7	78.72 ± 0.24	78.98 ± 0.27	79.18 ± 0.30	79.26 ± 0.23	78.66 ± 0.21	NS	NS	***	*	NS	*
pH	6.62 ± 0.01 c	6.65 ± 0.01 b	6.67 ± 0.01 a	6.67 ± 0.01	6.63 ± 0.01	**	*	NS	*	NS	NS
Acidity, g of lactic acid/100 mL	18.90 ± 0.11 a	18.52 ± 0.13 a	17.62 ± 0.14 b	19.03 ± 0.10	17.65 ± 0.10	***	***	*	NS	***	NS
Freezing point, °C	−0.55 ± 0.00 a	−0.55 ± 0.00 ab	−0.55 ± 0.00 b	−0.55 ± 0.00	−0.56 ± 0.00	*	*	**	NS	NS	NS

¹ Means in the same row followed by different letters differ significantly between SCC classes or between seasons. NS = not significant; * p < 0.05; ** p < 0.001; *** p < 0.0001. ² Somatic cell counts. ³ DIM_in classes <100;_≥100 <200; ≥200 <300; ≥300. ⁴ Score determined by the average of cows of each genetic group, were 1 = Jersey, 2 = crossbred (Holstein x Jersey, Holstein x Zebu, or Jersey x Zebu) and 3 = Holstein. ⁵ Parity, were 1 = 1 calving; 2 = 2 calving; 3 = 3 calving; 4 = ≥4 calving. ⁶ Energy-corrected milk. ⁷ Milk ethanol stability in Gay-Lussac.

). All milk attributes varied significantly across farms, DIM, parity, and genetic group (Table 1).

Table 2. Means ± mean square error and p-values of udder and teat attributes in lactating dairy cows according the SCC class (low, medium, and high) and season (cold and hot).

Variables	SCC Class			Season		p-Value 1
	Low (n = 392)	Medium (n = 297)	High (n = 249)	Cold (n = 431)	Hot (n = 500)	SCC 2	Season	DIM 3	Genetic Group 4	Parity 5	SCC x Season
Udder depth, cm	9.35 ± 0.25 a	8.19 ± 0.29 b	7.68 ± 0.32 b	8.80 ± 0.24	8.01 ± 0.22	**	NS	NS	***	***	NS
Udder clearance, cm	57.33 a	54.76 b	53.66 b	55.75	54.75	*	NS	NS	***	***	NS
Teat-end score 6	2.55 ± 0.04	2.66 ± 0.04	2.65 ± 0.05	2.66 ± 0.04	2.58 ± 0.04	NS	NS	NS	*	*	NS

¹ Means in the same row followed by different letters differ significantly between SCC classes or between seasons. NS = not significant; * p < 0.05; ** p < 0.001; *** p < 0.0001. ² Somatic cell count; ³ DIM_in classes < 100;_≥100 <200; ≥200 <300; ≥300. ⁴ Score determined by the average of animals of each genetic group, were 1 = Jersey, 2 = crossbred (Holstein x Jersey, Holstein x Zebu, or Jersey x Zebu) and 3 = Holstein. ⁵ Parity, were 1 = 1 calving; 2 = 2 calving; 3 = 3 calving; 4 = ≥ 4 calving. Score from 1 to 4, where 1 = clean and 4 = very dirty. ⁶ Teat-end score, rated from 1 to 4, where 1 = teat-end without ring formation and 4 = teat-end with a very rough ring; values represent the mean of the teats.

describes the characteristics of the cows in each SCC class and season. Based on the average values of udder depth and udder clearance, it was observed that the udders were closer to the hock and the ground in cows from the medium to high SCC classes (p < 0.05). Furthermore, a progressive reduction in udder depth and udder clearance was observed with advancing parity (p < 0.05).

The frequency of cows with hyperkeratosis (teat-end scores 3 and 4) also did not differ between SCC levels in the chi-square test (p < 0.05), with 56.27% of cows having hyperkeratosis in at least one teat. Although a significant effect of DIM was observed on the mean of the teat-end score, we did not consider this relationship to be scientifically relevant (Table 2).

Udder and leg cleanliness scores according to the SCC class and season were analyzed using the nonparametric Kruskal–Wallis test, and the results are presented below. The median leg cleanliness score was 3.0, regardless of the SCC class (p = 0.8660). Regarding the cold and hot seasons, the median leg cleanliness scores were 4.0 and 3.00, respectively (p < 0.0001). For udder cleanliness, the median score was 1.0, regardless of the SCC class (p = 0.1960). During both the cold and hot seasons, the median udder cleanliness score remained 1.0 (p = 0.5900).

There was an interaction between SCC classes and season for MY, fat concentration, fat yield, and MES (p < 0.05) (

Table 3. Interaction between SCC (low, medium, and high SCC) and different seasons (hot and cold) for milk characteristics.

SCC 1	Milk Yield, kg/Day				Fat Yield, kg/Day
	Cold Season	Hot Season	Average	p-Value	Cold Season	Hot Season	Average	p-Value
Low	16.69 Aa	15.42 Aa	16.05	0.364	0.75 Aa	0.61 Ab	0.68	<0.0001
Medium	16.37 Aa	14.01 Bb	15.19	0.005	0.71 Aa	0.59 Ab	0.65	0.001
High	14.48 Ba	13.98 Ba	14.23	0.982	0.62 Ba	0.57 Aa	0.59	0.530
Average	15.85	14.47			0.69	0.59
SCC 1	Fat, g/100 g				MES, °GL
	Cold Season	Hot Season	Average	p-Value	Cold Season	Hot Season	Average	p-Value
Low	4.62 Aa	4.04 Ab	4.33	<0.0001	78.42 Aa	79.00 Aa	78.71	0.975
Medium	4.57 Aa	4.26 Aa	4.41	0.077	79.46 Aa	78.49 Aa	78.97	0.836
High	4.46 Aa	4.14 Aa	4.30	0.154	79.88 Aa	78.47 Aa	78.47	0.573
Average	4.55	4.14			79.25	78.65

¹ Somatic cell count; Different capital letters in columns and lowercase letters in rows differ from each other (p < 0.05).

). Despite the significant interaction for MES, no significant differences were observed between the means (p > 0.05). Cows without subclinical mastitis produced more milk than cows with subclinical mastitis (p < 0.05) (Table 1 and Table 3). The highest ECM was found in the low and medium SCC groups and during the cold season (p < 0.05). Regarding the physicochemical attributes of milk, higher concentrations of protein and casein in milk were estimated for the high and medium SCC classes (p < 0.05). However, the daily protein yield and fat yield were higher in healthy cows than in high SCC cows. Lactose and chlorine concentrations were higher in the low SCC, followed by the medium and high SCC classes (p < 0.05), with high differences between the SCC classes for both traits. FP and the titratable acidity had higher means in the low SCC class (p < 0.05), with inverse results for pH (p < 0.05). The lowest concentrations of TSE and DDE were found in the high SCC group (p < 0.05). Fat and MES were not affected by the SCC class (p > 0.05). Although the casein-to-protein ratio varied between SCC classes (p < 0.05), the difference between a low and high SCC was less than 0.1 units, meaning that the values had low variability (ranging from 0.766 to 0.811), but were still within the normal range.

There was a lower ECM, acidity, fat, DDE, TSE concentration, casein-to-protein ratio and fat-to-protein ratio, and protein and lactose yield in the hot season, regardless of the SCC class.

The factor analysis formed six factors, explaining 63.2% of the total variation (

Table 4. Factor loadings, communalities, and percentage of variance related to milk yield and quality, herd characteristics, and udder conformation of dairy cows.

Variables	Factor 1	Factor 2	Factor 3	Factor 4	Factor 5	Factor 6	Communalities
	Milk Quality	Effect of Days in Milk	Cleanliness	Udder Conformation	Season Effect	Teat Condition
LnSCC 1	0.70250	0.17210	−0.03600	−0.05948	−0.00483	0.13426	53.30
Chloride content, g/L	0.82172	−0.12957	0.03988	−0.08546	−0.00279	−0.13733	77.47
Lactose, g/100 g	−0.86729	−0.17001	0.00025	0.01903	0.00133	−0.06414	78.86
Acidity, g of lactic acid/100 mL	−0.60990	0.47024	0.00614	0.13372	0.02945	0.01018	60.85
Fat, g/100 g	−0.05718	0.76189	−0.06905	−0.03837	0.02986	0.03934	57.09
Protein, g/100 g	−0.00232	0.88119	−0.06398	−0.03215	−0.08422	0.04715	78.72
Days in milk, days	0.32339	0.56218	0.17571	0.11449	0.06450	0.05767	46.20
Genetic group 2	0.12963	−0.37144	0.54767	0.22259	0.02739	0.04105	50.37
Cleanliness of the legs 3	−0.03251	0.07361	0.83715	0.03959	0.04494	−0.02409	71.49
Cleanliness of the udder 3	−0.02880	0.00137	0.75355	−0.17141	−0.21471	0.00652	60.41
Parity	0.15153	−0.11185	0.05518	−0.73815	−0.01320	−0.15872	66.09
Udder depth, cm	−0.07433	−0.10806	0.05552	0.87019	−0.04227	−0.12570	80.10
Climatic season 4	0.09774	−0.17952	−0.22965	−0.00282	−0.75648	0.14580	70.13
Freezing point, °C	0.09472	−0.31623	−0.10675	0.18320	0.55850	0.25745	55.98
Casein-to-protein ratio, g/100 g	0.00654	−0.06010	−0.29940	−0.14075	0.69339	−0.07987	53.82
Hyperkeratosis 5	0.15026	0.25895	0.04370	−0.00713	0.05348	0.78865	60.45
MES, °GL 6	−0.10358	−0.19094	−0.08812	0.06534	−0.24363	0.61944	52.17
Milk yield, kg/day	−0.31071	−0.27542	0.23046	−0.31915	0.17284	0.37696	64.00
Explained variance	17.96%	13.85%	9.88%	8.00%	7.45%	6.05%

¹ log-transformed somatic cell count (LnSCC = log2(SCC/100,000) + 3); ² Score determined by the average of animals of each genetic group, were 1 = Jersey, 2 = crossbred (Holstein x Jersey, Holstein x Zebu, or Jersey x Zebu) and 3 = Holstein; ³ Score from 1 to 4, where 1 = clean and 4 = very dirty; ⁴ Cold season (1) or hot season (2); ⁵ Teat-end score 3 + 4; ⁶ Milk ethanol stability in Gay-Lussac.

). Factor 1 explained 17.96% of the variation, in which lactose concentration was positively associated with acidity and negatively related to the SCC and chlorine content. Factor 2 explained 13.85% of the variance and showed that protein concentration was positively related to DIM, acidity, and milk fat concentration. Factor 3 explained 9.88%, where the genetic group and leg and udder cleanliness were positively associated. Factor 4 explained 8.00% of the variation, in which udder depth was negatively related to parity. Factor 5 explained 7.45%, in which the season was negatively associated with the fat percentage and casein-to-protein ratio of milk. The sixth factor explained 6.05% of the variance, where an increase in the hyperkeratosis score was positively related to the MES and MY. However, the MY of the cows did not have a factor loading greater than 0.4 in any of the factors, but it did show a relationship with lower values. This relationship was negative with the SCC (Factor 1) and positive with the increase in parity (Factor 4) and hyperkeratosis (Factor 6).

4. Discussion

Grouping dairy cows into low (less than or equal to 200,000 cells/mL), medium (between 200,001 and 615,000 cells/mL), and high (above 615,000 cells/mL) SCC classes allowed us to identify key traits associated with an elevated SCC. The influence of advanced DIM and higher parity on an increased SCC is well documented [6], and our statistical model reinforced their predictive value for milk quality and udder health. Particularly, cows with higher parity exhibited increased teat-end scores, highlighting the need to monitor these variables for more effective herd management.

Our findings align with previous research showing that udder conformation (closer to the hock or ground) [6], higher parity [14,32], and longer lactation periods [32] elevate the SCC and mastitis risk. These factors—combined with breed anatomy, milk yield potential, and immune adaptation—affect resilience to intramammary infection.

The positioning of the udder at or below the hock may reflect increased age, parity, and milk production [24]. Furthermore, these udders are more likely to become dirty due to their proximity to the ground, which may contribute to intramammary infections. A greater opportunity for contamination due to the conformation of the mammary gland, which affects the risk of new mastitis occurrences, has been demonstrated by Cardozo et al. [6]. In our study, there was no significant relationship between dirtiness and udder depth. Although the proximity of the udder to the ground did not favor an increase in udder dirtiness, udder depth was associated with the group of cows with a high SCC. As the distance between the udder and the hock decreased, the SCC increased.

Although hyperkeratosis was not statistically related to the SCC in our herd, international studies confirm it as a risk factor, especially when milking protocols are suboptimal (e.g., overmilking, poor vacuum control) and when hygiene is compromised. Teat-end mastitis can have multiple causes; thus, we believe that hyperkeratosis was not a relevant factor on these farms.

The teat-end should be smooth and free from lesions [27] to provide the necessary protection and prevent the entry of mastitis-causing pathogens. However, milking practices or equipment conditions can lead to hyperkeratosis [27], which impairs the sphincter’s full closure [33]. When the teat-end score exceeds 3, hyperkeratosis increases the risk of new cases of subclinical mastitis in the herd [6]. Pantoja et al. [34] reported that the interpretation of associations between hyperkeratosis and mastitis could be confounded by other factors, particularly parity, when they are not considered in the analysis.

Although this study did not focus on the risks associated with increased teat-end scores, several factors have been linked to the development of hyperkeratosis, such as improper pulsation ratios, overmilking, teat-end congestion, and removal of the milking unit without cutting off the vacuum [33]. These conditions can compromise teat tissue integrity, predisposing cows to chronic udder lesions. Additionally, frequent milking increases the chance of exposure to contagious pathogens [32], especially when equipment is not functioning properly. Therefore, regular inspection and maintenance of the milking system are essential measures to minimize the risk of chronic mastitis and promote udder health [6].

Cardozo et al. [6] reported a positive relationship between high milk production and hyperkeratosis. In contrast, our study was conducted on farms where cows had a relatively low MY. We observed only a weak association between hyperkeratosis and milk production, which diverges from the expectation of a stronger relationship in this group of dairy cows. Despite the low average MY, 57% of the evaluated cows presented some degree of hyperkeratosis, underscoring the relevance of monitoring teat-end health even in less intensive dairy production systems.

In our study, the daily MY was significantly lower in cows with subclinical mastitis during the hot season. A reduced MY is commonly observed in cows with a high SCC, as well as in animals exposed to nutritional deficits [7,8,13], advanced DIM, and higher parity [6,8]. Consistent with previous findings [7,13], the highest MY in our study was recorded in cows without subclinical mastitis [7,13]. According to previous research, daily milk losses begin when the log-transformed SCC (LnSCC) reaches 2.52 ± 0.63, and this loss is further aggravated by increased parity [8].

The effect of DIM on the lactation curve is well documented [8]. In our study, the daily MY progressively declined with advancing lactation, with average values of 17.77, 16.41, 13.87, and 12.59 kg/day observed in cows with DIM of <100, 100–200, 200–300, and >300 days, respectively.

The milk fat content was not affected by the SCC class, which aligns with findings from previous research [13]. However, the chloride and lactose contents were substantially influenced by SCC. The chloride content increased proportionally with SCC, corroborating previous findings [35,36,37]. Furthermore, the lactose content was negatively associated with the SCC, supporting the results of Alessio et al. [14,38]. Other studies have also reported reduced lactose levels in milk as the SCC increases [7,9,13,39].

Mastitis increases the permeability of tight junctions, allowing lactose to leak from the milk into the bloodstream [39]. In addition, inflammation and infection decrease the number of secretory cells [40]. Lactose plays a crucial role in maintaining osmotic balance between the blood and alveolar lumen [38]; therefore, a reduction in milk lactose is usually accompanied by an increase in blood-derived electrolytes [34]. The increased permeability of alveolar membranes enhances the transfer of citrate, bicarbonate, sodium, and chloride from the blood into the milk, while lowering concentrations of lactose, calcium, and potassium [36].

To date, this study represents one of the most comprehensive investigations of milk chloride content in lactating cows, encompassing a sample of 931 observations. Factorial analysis revealed a positive association between SCC and chloride levels, consistent with the conclusions of Gargouri et al. [18]. In Brazil, a qualitative assessment of chloride is required by law to detect milk adulteration by density [10]. However, our findings suggest that false positives may occur in the absence of adulteration, influenced by individual cow characteristics, diurnal variation, and SCC levels [41].

Increasing parity is known to be associated with a higher SCC. Accordingly, we observed a progressive increase in milk chloride concentration across lactations, with mean values of 0.87, 0.95, 0.97, 1.00, and above 1.00 g/L for the 1st, 2nd, 3rd, 4th, and >4th lactations, respectively. A similar association with DIM was also expected; however, no significant relationship was observed (p > 0.05). Gargouri et al. [18] classified chloride levels of <1.10, between 1.10 and 1.21, and >1.21 g/L with milk SCC ≤ 400,000, >400,000 to ≤800,000, and ≥800,000 cells/mL, respectively. Our study used narrower ranges, resulting in lower average chloride contents, as follows: 0.85 g/L for a low SCC, 0.93 g/L for a medium SCC, and 1.07 g/L for a high SCC.

Previous studies have reported that cows with an elevated SCC tend to have higher milk protein content [13,37]. Additionally, it was found that both the protein and casein content were higher in classes with a high or medium SCC. These findings suggest a potential correlation between mammary gland health and milk composition, particularly protein and casein levels. However, this relationship is also influenced by factors such as nutrition, genetics, and management practices. Despite a lower protein content in the low SCC class, these cows typically produced more milk, resulting in a dilution effect.

The higher casein-to-protein ratio in the low SCC class may indicate reduced milk proteolysis and a lower serum protein content [13]. Ogola et al. [36] found that the SCC was negatively correlated with the casein-to-protein ratio (r = −0.83, p < 0.05) and positively correlated with non-casein nitrogen (r = 0.70, p < 0.05) [36].

Milk collected during the hot season showed lower levels of fat, protein, lactose, casein, DDE, TSE, fat–protein ratio, casein-to-protein ratio, and protein yield. It is well established that seasonal variations and dietary factors strongly influenced milk component synthesis. Bernabucci et al. [19] observed similar season patterns, reporting significant reductions in milk protein (summer = 3.29% and winter = 3.5%), TSE (summer = 11.91% and winter = 12.58%), DDE (summer = 8.75% and winter = 9.00%), and fat (summer = 3.20% and winter = 3.8%). Although they did not detail the dietary composition, the authors suggested that a reduction in protein and casein synthesis could result from dietary constraints. The phosphorylation of caseins requires γ-phosphate from ATP, and during heat stress, reduced feed intake combined with increased maintenance requirements may lead to a negative energy balance. This supports the hypothesis that lower α- and β-casein levels during the hot season are related to limited dietary energy and protein, which impairs casein synthesis and alters the casein ratio.

Milk acidity and pH reflect the natural balance of raw milk, with an acceptable acidity ranging between 0.14 and 0.18 g of lactic acid/100 mL, according to Brazilian standards [10]. In this study, a higher SCC was associated with a lower acidity and higher pH, although both remained within regulatory limits [10]. Alhussien et al. [37] observed a slight increase in pH (6.61 vs. 6.63) in cows with subclinical mastitis. In our results, the milk pH increased from 6.62 to 6.67 between low and high SCC samples, confirming the findings of Alhussien et al. [37].

The hot season also influenced the acidity, pH, and FP of milk. We observed a lower acidity and higher pH, consistent with Bernabucci et al. [42]. These changes are likely due to reduced α- and β-casein levels, which are rich in phosphate groups and acidic components of the casein micelle [19,42]. Although we did not assess individual casein fractions, the total casein content was lower during the hot season. Although the FP remained within Brazilian regulatory limits [10], this parameter varied across SCC classes and seasons. Milk from medium and high SCC cows and samples collected during the hot season showed a lower FP, possibly reflecting higher solute concentrations and reduced water content. Although the MY decreased in these groups, this was not always mirrored in the TSE concentration.

To the best of our knowledge, this is the first Brazilian study to evaluate the FP of milk from pasture-based cows under different SCC levels and seasonal conditions. Previous research found a negative correlation between the FP and SCC (r = −0.36) in Holsteins fed total mixed rations [15]. Conversely, Kędzierska-Matysek et al. [16], studying confined cows in different seasons, found no SCC-related differences in the FP.

In Minas Gerais, Brazil, Oliveira et al. [43] analyzed bulk tank milk and reported a lower acidity and higher FP during winter, while the lowest MES values were found in late winter, corresponding to the dry-to-rainy season transition. In our study, no significant relationships were found between the FP and cow characteristics in factorial analysis, although other studies report associations with parity and DIM [16,40].

Finally, the FP was positively correlated with the casein-to-protein ratio, consistent with findings in Holstein and Simmental cows (r = 0.36 and r = 0.42, respectively) [15]. Variations in the FP likely reflect differences in milk protein, lactose, chloride, and water-soluble mineral concentrations (e.g., Ca, K, Mg), which are affected by the SCC and season [16,40].

5. Conclusions

Cows with a high SCC exhibited lower daily MY, reduced lactose content, and lower concentrations of total solids and defatted dry extract in the milk, regardless of the climatic season. Although the effect of the SCC on milk production, composition, and physical attributes was not dependent on the climatic season, the hot season negatively affected the synthesis of milk components, their yields, and physicochemical quality. In addition, cows with a high SCC produced milk with elevated chloride levels, increasing the risk of rejection by the dairy industry due to regulatory quality standards. Independent of the genetic group, selecting dairy cows with more favorable udder conformation—particularly shallower udders—is recommended, as deeper udders are associated with greater susceptibility to subclinical mastitis.