Compression of VK1-4 Pharma Grade API as a neat oil is not practicable on rotary tablet presses because the active ingredient is a lipophilic liquid at ambient temperature. A direct compression route is nevertheless feasible when the API is first adsorbed onto a porous excipient. In a 600 L high-shear mixer fitted with a bottom-drive impeller and side chopper, a premix of VK1-4 and colloidal silicon dioxide at a ratio of 1:2 w/w is processed at an impeller tip speed of 6–8 m/s and a chopper speed of 1800–2400 rpm for approximately 5–8 minutes. The resultant powder is discharged at a product temperature not exceeding 30°C, sieved through a 0.5 mm screen, and blended in a 1,000 L bin blender at 12 rpm for 20–30 minutes with microcrystalline cellulose, spray-dried lactose monohydrate, croscarmellose sodium, and magnesium stearate. Final blend bulk density is typically 0.45–0.55 g/mL, and the blend should demonstrate a Carr index below 25 and a Hausner ratio below 1.25 to allow uniform feed on a high-speed rotary press. Tablets containing 0.1–5 mg of VK1-4 are compressed on a 32-station rotary press using B-tooling at main compression forces of 8–15 kN, precompression forces of 3–5 kN, and turret speeds of 40–80 rpm. Content uniformity according to USP <905> should be evaluated at full press speed; the acceptance value is 15.0, but production-control limits of 10.0–12.0 are applied to manage segregation risk. Hardness is monitored on a Schleuniger tester at 50–90 N, and friability is controlled to not more than 1.0% per USP <1216>. Dissolution testing with USP <711> Apparatus 2 at 75 rpm in 900 mL of pH 6.8 phosphate buffer containing 0.5% w/v sodium lauryl sulfate at 37.0 ± 0.5°C is used; a Q value of 75% at 30 minutes is typically applied for low-dose phytomenadione tablets. The critical boundary for direct compression is residual moisture: when ambient relative humidity exceeds 60%, colloidal silicon dioxide adsorbs moisture and loses oil-binding capacity, so excipients should be pre-dried at 60°C for 4 hours before compounding.

What Governs Content Uniformity During High-Speed Capsule Filling of Adsorbed Phytomenadione?

Powder-filled hard capsule operations using VK1-4 require a different control strategy from tablet compression because the dosing mechanism on a dosator or tamping-pin capsule filler is sensitive to blend compressibility. In a tamping-pin machine equivalent to a GKF-type filler, the adsorbed powder should be granulated or densified to a bulk density of 0.50–0.60 g/mL before filling. Fill weight variation across 10 sample capsules should be controlled to relative standard deviation ≤ 2.0%; content uniformity per USP <905> should be measured at both the beginning and end of the fill run, and the variance between stations should not exceed 3.0% RSD. Capsule size 3 or 4 hard gelatin capsules are normally suitable for doses up to 5 mg; for higher-dose formulations, microcrystalline cellulose may be partly replaced with dibasic calcium phosphate anhydrous to increase tap density. Dissolution failures in capsules are generally caused by over-compaction of the powder slug inside the tamping pin. Fill weight control should therefore be coupled with slug porosity measurement: a target slug porosity of 0.25–0.35 and a compression stress of 0.5–1.2 MPa at the dosator tip prevent retarded release. At relative humidity above 55%, hard gelatin capsules soften and the powder slug may adhere to the pin; dehumidification to 40–50% RH and a filling room temperature ≤ 25°C are required.

Wet granulation becomes the preferred route when the target dose is 0.05 mg or below, because the direct adsorption premix cannot carry the API through a 20-fold dilution without segregation. VK1-4 is dissolved in ethanol or polysorbate 80 and sprayed onto a mannitol/starch substrate in a fluid-bed granulator. In a top-spray unit with inlet air temperature 45–60°C, product temperature 25–32°C, and spray rate 80–180 g/min, the solution is sprayed at atomization pressure 2.0–2.4 bar. Endpoint is determined by loss-on-drying 1.0–2.5% w/w and by visual absence of overwetted material at the bowl wall. The dried granules are milled through a 0.8 mm sieve; granule size distribution should have a d50 of 120–220 µm and more than 90% between 75 µm and 500 µm. Blend lubrication with 0.5–1.0% w/w magnesium stearate should not exceed 30 minutes because phytomenadione adsorbed on hydrophilic carriers can exhibit delayed dissolution after hydrophobic overlay. Residual ethanol is controlled under ICH Q3C Class 3 limits; not more than 5,000 ppm ethanol is typically applied for granule intermediates. Finished sachet or stick-pack granules are filled into triplex aluminum foil that blocks light below 0.5% transmission and provides moisture vapor transmission below 0.1 g/m²/day.

When Soft-Gelatin Capsules Are Used, the Fill Matrix Determines Isomer Stability and Shell Compatibility

Soft-gelatin capsule processing of VK1-4 avoids powder compressibility problems but introduces fill matrix and shell interaction risks. The API is diluted in medium-chain triglycerides or refined soybean oil at 5–20% w/w; butylated hydroxytoluene or alpha-tocopherol is added at 0.01–0.05% w/w to limit peroxide formation, and the fill is inerted with nitrogen to headspace oxygen below 2.0% v/v. Fill viscosity measured at 25°C should be 200–800 mPa·s to maintain volumetric dosing accuracy on rotary die encapsulation machines. Gelatin shell moisture content must stay within 35–45% of equilibrium moisture to prevent brittle shells; drying after encapsulation is carried out at 18–22°C and 20–30% RH for 24–72 hours depending on shell thickness. Migration of glycerin from the shell into the lipid fill can carry water and facilitate trans-to-cis isomerization; therefore the fill should be pre-dried with sodium sulfate, and the glycerin-to-gelatin ratio should be kept at 0.5:1 to 0.8:1. Softgel shells provide ultraviolet protection, but visible-light-protected cartons still apply. Dissolution testing uses USP <711> with a medium that may include sodium lauryl sulfate at 0.5% w/v; gelatin crosslinking should be excluded by adding pepsin when indicated by pellicle formation.

Sterile Injectable Emulsions Are Governed by Droplet Size and Terminal Sterilization Feasibility

Parenteral delivery of VK1-4 is possible through a mixed-micelle solution or an oil-in-water emulsion. Mixed-micelle concentrates are prepared by dissolving the API in soybean oil and lecithin, then dispersing with sodium glycocholate and polysorbate 80 in water for injection to a final dose of 1–10 mg/mL. The emulsion route is preferred for hemodynamic safety because the lipophilic API partitions into the oil core and reduces free-drug precipitation. Pre-emulsion is formed in a high-shear mixer at 10,000–18,000 rpm for 10–15 minutes, then homogenized through a two-stage high-pressure homogenizer at 10,000–15,000 psi with a second-stage pressure of 1,500–2,500 psi for 6–12 passes. Droplet size is monitored by dynamic light scattering; the mean droplet diameter should be 200–350 nm and the 90th percentile should remain below 600 nm, as larger droplets increase the risk of emulsion cracking during terminal sterilization. Terminal sterilization in a rotating autoclave at 121°C for 15 minutes is appropriate only if droplet size stability is verified; phytomenadione degrades less than 2.0% after autoclaving when nitrogen overlay is used and headspace oxygen remains below 1.0% v/v. Filter sterilisation is not generally feasible for lipid emulsion droplets above 220 nm; therefore terminal steam sterilisation is the default if formulation and droplet size permit. The final emulsion is filled into amber borosilicate glass vials under nitrogen and stored at 2–8°C. Particulate matter is controlled by USP <788> Method 1: not more than 6,000 particles ≥10 µm and 600 particles ≥25 µm per container. Endotoxin is tested to Ph. Eur. 2.6.14; the limit should not exceed 0.5 EU/mg of phytomenadione for intravenous administration.

Oral liquid dosage forms of VK1-4 are prepared as oil-based drops or aqueous mixed-micelle solutions for neonatal prophylactic and ambulatory anticoagulant reversal protocols. For oil-based drops, the API is diluted in medium-chain triglycerides to a concentration of 2 mg/mL and filled into amber glass bottles with a calibrated dropper tip. The dose is typically 0.1–0.5 mL depending on the clinical protocol; the fill volume accuracy of the dropper should be validated at ±0.01 mL. Aqueous mixed-micelle solutions use polysorbate 20 or polysorbate 80 at 2–4% w/w and may include sorbitol as an isotonicity agent; the finished pH is adjusted to 5.0–7.0 with citrate buffer. Physical instability presents as turbidity if the micelle solubilization capacity is exceeded, and the formulation must remain clear after storage at 25°C/60% RH for 6 months and at 2–8°C for 24 months. The light-sensitivity of VK1-4 is more severe in dilute aqueous presentation than in oil: amber glass bottles with visible light transmission below 10% at 450 nm and secondary cartons are mandatory. Microbiological quality follows USP <1111> and Ph. Eur. 5.1.4; oral liquids should be preserved with 0.1% sodium benzoate or 0.05% potassium sorbate when multi-dose containers are used.

Lyophilized Cyclodextrin Complexes for Reconstituted Parenteral Use

Complexation of VK1-4 with hydroxypropyl-beta-cyclodextrin improves aqueous solubility and enables lyophilization. In a molar ratio of 1:20 to 1:50, the complex is prepared by mixing the oily API with aqueous HP-β-CD under nitrogen, then homogenizing at 8,000–12,000 rpm. The resulting solution is filtered through a 0.2 µm sterilizing-grade filter because droplet-free inclusion complexes can pass; lyophilization is performed at a shelf temperature of -45°C for freezing, primary drying at -20°C and 80 mTorr for 24–48 hours, and secondary drying at 25°C for 6–12 hours. The cake should retain ≤ 1.0% w/w moisture and reconstitute in 30 seconds with water for injection. The main limitation is limited drug loading; if the cyclodextrin concentration exceeds 20% w/v, the cake becomes collapsed, and the residual moisture increases to 2–3% w/w. This route is used where organic-solvent-based injectable emulsions are not suitable in neonatal populations.