During the post-weaning and grow-finish stages, standardized Kushen powder (matrine 10.0% ± 0.5% by HPLC) is incorporated into dry feed premixes for swine enteric health management. A production-scale batch is prepared in a 250 L double-ribbon horizontal blender charged to 60% working volume. The batch comprises 18.0 kg of corncob meal screened through 40–60 mesh, 1.2 kg of standardized Kushen powder, and 0.8 kg of precipitated silica. The silica is pre-blended with the botanical extract for 6 min at 18 rpm before the carrier is added; this step prevents extract agglomeration and maintains bulk flow. The full blend is mixed for 15 min at 18 rpm, discharged through a 20-mesh security screen, and sampled for homogeneity according to VICH GL18 statistical principles. A coefficient of variation below 5.0% is required; if sampling shows CV above 7.5% after 12 min, mixing is extended in 3-minute increments but must not exceed 25 min total because over-mixing can induce segregation of the high-surface-area botanical fraction. Prior to discharge, 0.2% w/w food-grade mineral oil is sprayed through an atomizer during the final 3 min to suppress dust. Finished premix moisture is controlled below 9.0% by USP <921> loss-on-drying, and the material is packed in 25 kg multi-wall paper bags with a 0.08 mm polyethylene inner liner. Storage conditions are maintained at or below 25°C and 60% relative humidity. Alkaline feed ingredients and urea-based buffers must be segregated from this premix, because localized pH above 7.5 can shift quinolizidine alkaloids toward free-base forms and reduce blend stability.

For pelleted complete feeds, the premix is metered at 0.5–1.5 kg/t depending on the veterinarian-directed matrine target in the finished ration. Conditioning and pelleting steps require a strict temperature ceiling: conditioner steam settings that produce mash temperatures above 85°C are avoided because published degradation data for this specific extract configuration are limited above that threshold. A validated production window of 78–82°C with a conditioner residence time of 30–45 s is used where pellet durability permits. Conditioning moisture is kept below 16.5%; higher free-water levels at the die lips promote extract migration toward the pellet surface, producing uneven distribution and visible specking. The terminal feed form is a 3.5–4.5 mm pellet for grow-finish swine, and finished feed samples are retained for 12 months to support batch traceability under ISO 22005 or equivalent feed-chain documentation.

Why Does pH Shift After Reconstitution in Poultry Drinking Water Systems?

A water-soluble powder dosage form for broiler and layer drinking water is formulated as a 100 g sachet containing 20 g standardized Kushen extract, 15 g anhydrous citric acid, 10 g sodium citrate dihydrate, and 55 g dextrose monohydrate. The citrate buffer holds reconstituted water pH at 4.8–5.2, which maintains matrine in a protonated, water-soluble state. However, in production houses drawing hard water with bicarbonate alkalinity above 250 mg/L, the reconstituted pH can drift to 7.4–7.8 within 6 h. In such circuits the sachet rate is increased to 1.5 g/L or the stock solution is supplemented with 0.3 g/L sodium bisulfate. Residual free chlorine above 2.0 mg/L in nipple drinker circuits oxidizes the quinolizidine alkaloids; a bypass tank with sodium thiosulfate at 25 mg/L is therefore required before the product is introduced. Reconstituted solutions are typically administered at 1 g/L for 5–7 days and are prepared fresh every 24 h because no preservative system is included in this dry powder form.

Manufacturing of the soluble powder uses co-milled extract and dextrose monohydrate passed through a 60-mesh screen, followed by blending in a 200 L V-cone blender at 25 rpm for 20 min. Sachet filling is conducted at 35% relative humidity or below to prevent moisture pickup and caking. Release tests include moisture by USP <921> with a limit of ≤5.0%, and uniformity of dosage units by USP <905> with an acceptance value of ≤15.0. The terminal configuration is a foil laminate sachet in 100 g, 500 g, or 1 kg formats. Simultaneous acidifier dosing below pH 3.0 is not recommended, because the buffer capacity of the citrate system is exceeded and free alkaloid base may precipitate in the drinker line.

Injectable Sterility Assurance and Endotoxin Control for Subcutaneous Matrine Formulations

Injectable processing of Kushen powder begins with a refined water-soluble extract standardized to 10 mg/mL matrine and 2 mg/mL oxymatrine in the final aqueous solution. The extract is dissolved in Water for Injection at 35°C, and the pH is adjusted to 6.4–6.6 with 0.1 M hydrochloric acid or sodium hydroxide. Activated carbon at 0.1% w/v is added at 60°C for 15 min to reduce pyrogens and colored degradation products; the carbon is then removed through a 0.45 µm polyethersulfone membrane. The sterile filtration step uses a 0.22 µm PVDF filter cartridge under gradient pressure not exceeding 2.0 bar. Sodium chloride is added to achieve an osmolality of 280–300 mOsmol/kg. The solution is filled into 10 mL amber Type I glass vials and closed with fluoropolymer-coated butyl elastomer stoppers under Grade A air supply. If heat stability studies confirm greater than 95% matrine recovery, terminal steam sterilization at 121°C for 15 min with an F0 of at least 8 is used; otherwise, the batch is released as an aseptically filtered product. Sterility is tested by USP <71>, bacterial endotoxins by USP <85> with a limit of ≤0.5 EU/mL, and particulate matter by USP <788> for small-volume injections. The vial must contain no more than 6000 particles ≥10 µm and no more than 600 particles ≥25 µm per container.

Storage below 2°C may generate low-temperature turbidity; containers should be warmed to 20°C and visually inspected before administration. The injection must not be mixed in the same syringe with alkalizing solutions such as 8.4% sodium bicarbonate, because precipitation of free-base alkaloids occurs above pH 7.5. Residual solvent levels are controlled under VICH GL18 with ICH Q3C Option 1 limits applied by the manufacturer. Batch records include the extract-to-solution mass ratio, final pH, filter integrity results, and endotoxin load before sterile filtration to document process consistency.

When Tablets Replace Drenching: Direct Compression Boundaries for Hygroscopic Extracts

A veterinary tablet containing 200 mg standardized Kushen extract per unit, equivalent to 20 mg matrine, is feasible by direct compression only when ambient relative humidity is held at 35% or below. The direct compression formula comprises 200 mg extract, 80 mg silicified microcrystalline cellulose, 60 mg lactose monohydrate, 8 mg crospovidone, 4 mg sodium starch glycolate, 2 mg colloidal silicon dioxide, and 2 mg magnesium stearate. The extract and silicon dioxide are pre-blended 1:1 for 5 min, passed through a 30-mesh screen, and then added to a bin blender with the remaining excipients. Disintegrant is incorporated after the primary blend, and lubricant is added for the final 3 min only. Compression proceeds on a 16-station rotary tablet press to a hardness of 5–8 kp, friability below 0.8% by USP <1216>, and disintegration below 15 min by USP <701>. If the processing suite exceeds 40% RH or the extract moisture exceeds 2.5%, direct compression is abandoned in favor of wet granulation with 5% povidone K30 in an isopropanol–water 70:30 mixture. Wet granulation dries at 55°C inlet air with product temperature maintained at 35°C, targeting an LOD of 2.0–2.5% by USP <921>.

Release testing for the finished tablet includes dissolution by USP <711> Apparatus II at 50 rpm in 0.1 M hydrochloric acid, with a Q value of 75% dissolved at 45 min. Content uniformity by USP <905> must show an acceptance value of ≤15.0. Aqueous film coating uses hypromellose 6 cps, PEG 400, and titanium dioxide to a 3% weight gain. The coated tablets are packed in 10×10 aluminum–PVC blister cards with desiccant sachets to limit moisture ingress.

Controlling Alkaloid Segregation During Capsule Filling for Low-Dose Veterinary Blends

Hard gelatin capsule filling for small-animal doses requires a granulated matrix rather than direct powder encapsulation, because standardized Kushen powder at low mass fractions segregates in the hopper due to differences in particle size and static charge. A size 3 capsule formulation with a total fill weight of 200 mg contains 50 mg standardized Kushen extract, 120 mg lactose monohydrate, 25 mg microcrystalline cellulose, 4 mg crospovidone, and 1 mg povidone K30 as binder. Wet granulation is performed in a high-shear granulator with 5% povidone K30 dissolved in an isopropanol–water 70:30 solvent system. The granulate is dried to an LOD of 2.0–3.0%, passed through a 20-mesh screen, and filled on a dosator-type capsule machine at 60,000 capsules/h. The filling room is maintained at ≤35% RH to prevent gelatin shell deformation and extraction of moisture by the botanical granulate. At higher humidity, static charge increases and hopper segregation can raise the coefficient of variation above 6.0%; in that state, content uniformity failures become detectable even before the acceptance value limit is reached.

The release specification for capsules includes content uniformity by USP <905> with an acceptance value of ≤15.0, and dissolution by USP <711> Apparatus I at 100 rpm in 900 mL water with a Q value of 80% at 60 min. Magnesium stearate is not included at levels above 1.0% w/w, because hydrophobic lubrication delays alkaloid release in the aqueous medium. The capsules are packaged in HDPE bottles with a 1 g silica gel desiccant canister and induction-sealed foil liner. Terminal product storage below 25°C and 60% RH is specified.

Granule Feed Top-Dressing: Moisture Activation and Flowability in Fluid-Bed Processing

A sugar-free granule top-dressing is produced by fluid-bed top spray granulation in a 5 kg batch unit fitted with a top spray insert, using inlet air at 70°C and product temperature controlled at 40–45°C. The matrix contains 70% mannitol, 20% pregelatinized starch, and 10% standardized Kushen extract. Binder is 2% povidone K30 in purified water, sprayed at 50 g/min with atomizing air at 1.5 bar. After drying to ≤3.0% LOD by USP <921>, the granules are screened to retain 0.5–1.0 mm particles at a minimum of 90%. Powder flow is characterized by Car’s index below 15 and Hausner ratio below 1.18, ensuring uniform metering from a graduated scoop. The terminal product is filled into 1 kg PET jars with foil induction seals and a desiccant canister. Because the granules are hygroscopic, the filling line is maintained at ≤35% RH.

This top-dressing form is applied over the daily ration rather than incorporated into the pellet matrix post-extrusion. Field handling must avoid fat-coated pelleted feed surfaces, because hydrophobic surface films reduce granule adhesion and create uneven intake. A metered scoop is supplied to standardize the dose, and the dosing instruction is printed on the label in mg/kg body weight. Production records include fluid-bed exhaust humidity, spray rate, and final particle-size distribution to document batch-to-batch consistency.

Does Oxymatrine Reach a Solubility Ceiling in Co-Solvent Oral Solutions?

An oral solution for swine and cattle is prepared as a co-solvent system to maintain clarity at higher botanical extract loads. A 100 mL formulation contains 10 g standardized Kushen extract, 20 mL propylene glycol, 10 mL glycerin, 0.1 g potassium sorbate, 0.1 g sodium benzoate, and purified water to volume. Citric acid is added to adjust pH to 4.5–5.5. The extract is dissolved in purified water at 35–40°C; the preservatives are separately dissolved in propylene glycol and added with stirring. The solution is cooled to 25°C, filtered through a 5 µm polypropylene cartridge, and filled into amber PET bottles with LDPE cap liners. Matrine solubility is adequate in the co-solvent system up to roughly 2.0% w/v under the stated pH range; above that concentration or above pH 7.2, turbidity and free-base precipitation may occur. Hard water with total hardness above 300 mg/L as CaCO₃ should not be used for dilution, because carbonate interactions can reduce clarity at the drinker trough.

The oral solution is stored below 25°C and protected from light. Freezing below 2°C may produce reversible low-temperature precipitates; the container is warmed to room temperature and shaken before administration. Release testing includes pH, HPLC assay, preservative content, and microbial limits by USP <61>/<62>. The terminal package is a 1 L amber PET bottle with a graduated dispensing chamber. The product label specifies a shake-before-use instruction and a 28-day use period after first opening.