The binary sterile mixture of ceftazidime and anhydrous sodium carbonate is received as a white to faintly yellow crystalline powder in double polyethylene-lined aluminium cans. In an injectable manufacturing facility operating under EU GMP Annex 1 (2022) and 21 CFR 210/211, the material is transferred into a closed isolator or RABS supplied with unidirectional airflow meeting ISO 14644-1 Class 5 at rest and in operation. The aseptic core is maintained at a differential pressure of at least 10 Pa relative to the surrounding ISO 7 background. Vials are Type I borosilicate glass, washed and depyrogenated by dry heat at 250°C for 45 min or equivalent; stoppers are siliconized chlorobutyl closures sterilized by gamma irradiation at 25 kGy or steam at 121°C for 20 min. The powder is filled using a gravimetric or vacuum-assisted auger filler with in-process gravimetric checks every 15 min; the target fill weight is adjusted to deliver 1 g of ceftazidime activity with approximately 118 mg of sodium carbonate per vial. Because the sodium carbonate component is hygroscopic and can generate carbon dioxide in humid air, the fill zone is maintained at a dew point below -30°C and relative humidity below 10% RH. After filling, the stoppers are seated and sealed with aluminium flip-off caps, and residual seal force is measured on a sample of vials. Released batches are tested for sterility, bacterial endotoxins, particulate matter, and water content; environmental monitoring during filling includes settle plates, contact plates, and active air sampling. Any action-level excursion triggers a batch sterility investigation under 21 CFR 211.192.

How Does Sodium Carbonate Govern Reconstitution pH and Carbon Dioxide Release in Aseptic Vials?

When 10 mL of sterile water for injection is added to a vial containing 1 g of ceftazidime with 118 mg of sodium carbonate, the powder dissolves and the carbonate species establishes a solution pH in the compendial range of 5.0–7.5. The theoretical carbonate load is 1.11 mmol, based on a sodium carbonate molecular weight of 105.99 g/mol; in 10 mL of water this corresponds to an initial carbonate concentration of 111 mmol/L before protonation by ceftazidime acidic groups and dissolved carbon dioxide. Below pH 5.0, the solubility of ceftazidime declines and the undissolved fraction may contribute to visible particulate matter; above pH 7.5, beta-lactam ring hydrolysis is accelerated, producing a yellow to amber discoloration and pyridine degradation products. The endpoint is judged by a clear to light-yellow solution without turbidity or gas pockets larger than 2 mm in the syringe. During automated reconstitution in hospital infusion devices, carbon dioxide generated from carbonate protonation can create back-pressure and foam; therefore, the vial is vented with a sterile needle before positive-pressure withdrawal. Solubility is influenced by the quality of water for injection: compendial water for injection with pH 5.0–7.0 and total organic carbon below 500 ppb avoids introducing acidic moieties that consume carbonate and lower pH. The pH is measured with a calibrated glass electrode according to USP <791>; temperature correction to 25°C is applied. For dilute intravenous solutions, the buffer capacity of the formulation is negligible once added to 50 mL or 100 mL of 0.9% sodium chloride; the resulting admixture pH is governed by the diluent and should be checked if infusion is prolonged beyond 24 h at 22–25°C.

Extemporaneous preparation of ceftazidime infusion solutions in hospital cleanrooms follows USP <797> or local pharmacopoeial compounding standards. The reconstituted vial is diluted to a final concentration typically between 1 mg/mL and 40 mg/mL in 0.9% sodium chloride, 5% dextrose, or Ringer's lactate; the choice of diluent affects the physical stability of the solution. In elastomeric infusion pumps, the solution may be exposed to 31–37°C near the patient's body for up to 24 h. Compatibility screening with vancomycin is negative: simultaneous administration through the same intravenous line without flushing has produced white precipitation at Y-site contact points. If aminoglycosides are prescribed concurrently, separate infusion lines or a 0.9% sodium chloride flush of at least 10 mL is used between agents. Particulate monitoring of compounded admixtures is performed with a light obscuration method per USP <788>; for small-volume containers the acceptance limits are 6000 particles per container at ≥10 µm and 600 particles per container at ≥25 µm, while a 100 mL infusion bag is evaluated against the large-volume limits of 25 particles/mL at ≥10 µm and 3 particles/mL at ≥25 µm. The pH of the final admixture is not buffered by the residual sodium carbonate at these dilutions; therefore, extended infusion protocols exceeding 24 h require stability data generated with the specific elastomeric device material. Published data for this specific ceftazidime–elastomeric device combination is limited.

Dry Granulation of Moisture-Sensitive Ceftazidime–Sodium Carbonate Blends for Experimental Low-Dose Tablets

Oral administration of ceftazidime is not a licensed indication, because the beta-lactam ring is hydrolyzed in gastric fluid and the compound shows negligible intestinal permeability. Nevertheless, formulation development for clinical pharmacology studies may require a tablet or capsule presentation; in such cases, the sterile ceftazidime–sodium carbonate blend is treated as a moisture-sensitive and low-density active ingredient. Direct compression is generally avoided because the blend has poor flow, with Carr index values frequently above 30% and Hausner ratios above 1.35 in unconditioned powder. Incoming particle size distribution is measured by laser diffraction; a D90 below 250 µm is required to avoid ribbon splitting during roller compaction. Dry granulation by roller compaction is preferred. A laboratory-scale roller compactor with roll force control from 5 kN/cm to 25 kN/cm, roll gap 1–3 mm, and an integrated oscillating granulator with 0.8 mm and 1.6 mm screens is used. The granulated material is blended with microcrystalline cellulose, crospovidone, and magnesium stearate in a bin blender; colloidal silicon dioxide at 0.5–1.0% w/w is added to reduce electrostatic adhesion. Compression is performed on a rotary tablet press with a compaction force of 8–18 kN and a target tablet hardness of 40–80 N for a 200 mg tablet. In-process tests include weight uniformity per USP <905> with an acceptance value not exceeding 15.0, disintegration per USP <701> in 0.1 N hydrochloric acid, and dissolution per USP <711> using 900 mL of 0.1 N hydrochloric acid at 37°C with paddle speed 50 rpm. Chemical stability in this medium is a major constraint: ceftazidime degrades rapidly below pH 2.0, so dissolution samples require immediate dilution into phosphate buffer pH 6.8 and HPLC analysis within 15 min. The sodium carbonate component raises the local pH during disintegration but does not protect the drug after the buffer capacity is exhausted. Published data for this specific configuration is limited.

Capsule filling of the sterile ceftazidime–sodium carbonate blend for phase I clinical trial supplies is carried out in a relative-humidity-controlled room because the sodium carbonate fraction will sorb moisture above 40% RH and cause sticky powder flow and gelatin shell softening. Hydroxypropyl methylcellulose capsules are preferred over hard gelatin when moisture sensitivity is a concern; the shell water content is specified below 10% w/w. A tamping-pin capsule filler with a powder bed height of 10–20 mm and tamping pin compression settings from 5 mm to 15 mm is used to achieve a reproducible fill weight of 100 mg of the blend into size 0 capsules. Blend uniformity is verified by sampling 10 locations across the powder bed and assaying ceftazidime by HPLC using USP <621> system suitability criteria for resolution and tailing. The sodium carbonate content is confirmed by acid–base titration with 0.5 N hydrochloric acid and methyl orange indicator; the acceptance range is set at 90.0–110.0% of the labeled amount. Dissolution testing in 900 mL of 0.1 N hydrochloric acid at 37°C shows complete dissolution of the capsule shell within 5 min, but ceftazidime degradation products appear rapidly; therefore, the dissolution profile is not considered a surrogate for oral bioavailability.

When Granule Reconstitution Is Evaluated Against pH-Dependent Bioavailability Limits

Granule presentations containing ceftazidime with sodium carbonate are evaluated for reconstituted suspensions only in settings where gastrointestinal stability can be modified by enteric coating or where the target species has a non-human gastric pH profile. The sodium carbonate component acts as a pH-raising excipient, but its buffering capacity is limited: 118 mg of sodium carbonate in 100 mL of water gives a carbonate concentration of approximately 11.1 mmol/L, which is insufficient to neutralize 500 mL of simulated gastric fluid at pH 1.5–3.5. Suspension development therefore uses a high-viscosity vehicle containing xanthan gum at 0.2–0.5% w/v and a preservative system that does not contain sodium bisulfite, because sulfite can accelerate beta-lactam degradation. Wet granulation with aqueous binder solutions is excluded due to carbon dioxide generation and poor granule strength. Instead, the active blend is compacted and the granules are coated with an aqueous methacrylic acid–ethyl acrylate copolymer dispersion to a weight gain of 15–20% in a fluid-bed coater with inlet air temperature 35–45°C and dew point below 10°C. Acid resistance is tested in 750 mL of 0.1 N hydrochloric acid for 2 h, followed by pH 6.8 phosphate buffer; ceftazidime release in the buffer stage must exceed 80% of the labeled amount within 45 min for the formulation to be considered a delayed-release granule. Because oral ceftazidime bioavailability remains negligible even with enteric protection in the absence of a permeation enhancer, published data for this specific configuration is limited.