SENSORY AND ANALYTICAL DETERMINATION OF OFF-FLAVOR COMPOUNDS IN WATER AND FARM RAISED CATFISH

Off-flavors in aquaculture are a global and persistent problem, which decreases product quality and increases production costs. Remedial treatments include salting, smoking, marinades and purging prior harvest .  In this report we used a sensory panel to show that consumers can perceive the earthy, musty, muddy aromas produced by the metabolites 2-methylisoborenol (2MIB) and geosmin in the low parts per trillion range in water. Due to the  presence of other  aroma producing compounds found in aquaculture products such as salmon, trout, or catfish, the off-flavor aromas are less noticeable and require higher levels of concentration for sensory detection. Using a forced triangle test (3-AFC), 3 out of 7 panelists were able to perceive a difference in a sample of catfish spiked with geosmin at a concentration of 100 parts per trillion,  where 5 of 7 panelists were able to perceive a difference in a sample spiked at a concentration of 400 parts per trillion. The odor was not necessarily classified as objectionable or identifiable , but  it  could be detected.  

Whereas off-flavor detection by sensory analysis is rapid and inexpensive, analytical detection can provide quantitation and improved sensitivity in most cases.  Instrumental analysis is  valuable for determining the efficacy of whether a remedial treatment such as purging has been effective. A pass/fail method for the screening of samples h as been developed to increase  sample  throughput at the expense of sensitivity.   The pass/fail method uses only 1 gram of tissue, which is placed in a 10 ml sample vial, with 5 ml of H2O and 3 g of NaCl.  The vial is then sonicated for 30 minutes and  analyzed using  solid phase microextraction gas chromatography mass spectrometry (SPME/GC/MS). Using th e  pass/fail method, 2MIB  and geosmin  in catfish  at concentrations on  a par with sensory thresholds can be detected.  A decrease in sensitivity  relative to the microwave distillation /SPME/GC/MS method  results from a  decrease in sample size (20:1).  Additionally some of the analyte remains in the sample  primarily as a result of the presence of lipid which in turn inhibits the quantification of the off-flavor analytes.  With limits of detection below  sensory thresholds for both analytes, t his technique is useful  for screening large numbers of samples for off-flavor.

Figure 1: Correct responses of seven panelists at different concentrations of geosmin in catfish.