HPLC-CAD Application Note for Tobramycin Assay


Tobramycin is an aminoglycoside antibiotic. Because aminoglycosides have no conjugated structure and weak UV absorption, routine direct UV detection is not ideal.
The 2020 Chinese Pharmacopoeia uses ELSD for related substances in tobramycin raw materials, while the assay item uses a microbiological method. Current USP standards use thin-layer chromatography for related substances and UV detection after derivatization for assay. In January 2023, the United States Pharmacopeia Commission issued a draft proposing HPLC-CAD for related substances and assay of tobramycin.
This application evaluates an Instrumax CAD method with an Agilent ZORBAX SB-C18 column for tobramycin assay. The method shows strong specificity, high sensitivity, good repeatability and good linearity.

Instrument: Agilent 1200 liquid chromatograph + Instrumax CADetector a1 Charged Aerosol Detector.
Reagents: pure water (Watson’s distilled water), acetonitrile (chromatographic grade), trifluoroacetic acid (analytical grade). Sample: tobramycin.
Column: Agilent ZORBAX SB-C18, 4.6 mm × 250 mm, 3.5 μm. Column temperature: 10 °C. Flow rate: 0.8 mL/min. Injection volume: 25 μL.
Mobile phase A: 0.3% TFA aqueous solution. Mobile phase B: 0.3% TFA in acetonitrile. Gradient elution.
CAD conditions: drift tube temperature 40 °C, nebulizing gas flow 3 L/min, corona current 1 μA, corona gas flow 1 L/min, gain 0.2.

Blank solvent: mobile phase A. Tobramycin reference-standard stock solution: 180.0 μg/mL.
Linearity solutions: 10.8, 14.4, 18.0, 21.6 and 27.0 μg/mL.
The blank does not interfere with the main peak, indicating good specificity.
Figure 1. Blank gradient (no injection). Figure 2. Blank solvent (mobile phase A). Figure 3. Tobramycin reference stock solution at 180.0 μg/mL.

For linearity solution 1 at 10.8 μg/mL, the USP theoretical plate number was 16600.5 (>3000), showing good column performance.
For the 18.0 μg/mL reference solution, five consecutive injections gave a peak-area RSD of 0.980% (<2.0%), indicating good precision.

Linearity solutions 1 to 5 were each injected once, and linear regression was performed using solution concentration and peak area. The linear correlation coefficient was R = 0.99918 (>0.995), showing good linearity.

The blank gradient had a gradient peak at 15–16 min, which did not affect the experiment. The blank solvent had four peaks at 2–5 min, far from the retention time of the tobramycin main peak, so they did not interfere with detection.
The results show that Instrumax CAD provides good detection, repeatability and linearity for tobramycin samples, meeting the needs of method evaluation and quality control.
Reference: USP-NF draft for comments, Tobramycin [S].

