Antimicrobial residues detected in marketed milk in urban and rural areas in Kenya

Abstract

Anti-microbial agents are widely used in Kenya for the treatment of diseases in both man and animals. Of major public health concern is the possible presence of anti-microbial residues in milk that is not withheld from human consumption or due to addition by market agents as preservative. This is aggravated by inadequate national anti-microbial residue surveillance and control programme in Kenya. A study was therefore conducted to establish the extent of contamination of marketed milk by anti-microbial residues in urban and rural areas in Kenya. Respondents were randomly selected and questionnaires completed in consumer households and market agents from Nairobi, Kiambu, Nakuru and Narok districts. A total of 916 raw and processed milk samples were collected seasonally between January 1999 and January 2000 from the four districts and analyzed for anti-microbial residues using CharmAIM96 test. Of the 916 milk samples analyzed, 348 samples were collected from consumer households while 458 samples were from informal market agents of various cadres. The remaining 110 samples screened were processed milk from Nairobi and Nakuru districts. Positive milk samples on Charm AIM96 were screened for tetracyclines and beta lactam antibiotics using Charm ROSA tests. Simple tests of association were first used to assess the relationship between the presence of anti-microbial residues in rural and urban households milk samples. A logistic regression model was thereafter fitted to evaluate risk factors associated with anti-microbial residue presence in marketed milk. Test agreement between the CharmAIM96 and Charm SL tests, for analysis of anti-microbial residues in milk was assessed. In addition, the Charm ROSA test was experimentally validated.

The Charm AIM96 test showed that 41(11.8%) and 25(5.4%) of milk samples from consumer households and market agents had anti-microbial residues above the FAO/WHO Codex MRL respectively. Nine out of 110(8.2%) pasteurized milk samples had resides above the FAO/WHO Codex-MRL. None of the consumer and market level milk samples was positive on the Charm ROSA tests. The proportions of rural and urban consumer household samples with anti-microbial residue above the FAO/WHO Codex-MRL were 18.7% and 5.1% respectively. This was a significant difference at 95% confidence level in proportions of rural and urban consumer household samples with detectable residues level (X2= 15.5: P value=0.000). The proportions with the residues above the FAO/WHO Codex-MRL decreased with increasing levels of milk bulking with small mobile traders, milk bars and milk shop kiosks having 8.5%, 6.4% and 4.2% respectively. Season, fanning systems, sales volume, and market access and channels were not associated with anti-microbial residue presence in marketed milk (P >0.05). Charm AIM96 and Chann ROSA tests detected penicillin G and oxy-tetracycline residue in all the eight milk samples collected up to 48 hours after treatment at level above 5ppb and 125ppb respectively. Chann ROSA tests detected oxy-tetracycline, penicillin G and amoxycillin at levels above 125ppb, 5ppb and 3.1ppb respectively. In conclusion, anti-microbial residues were more likely to have originated at farm level. The sensitivities of Charm SL tests obtained from this study were in agreement with what was indicated by Chann Sciences the manufacturers of Chann ROSA kit. Although Chann AIM96 and Chann ROSA tests detected penicillin G or oxy-tetracycline residues in post-treatment milk samples up to 72 hours, the

detection agreement between the two tests was low for samples collected beyond the third day after treatment, and inconclusive for field milk samples tested.