A practical, origin-by-origin playbook for managing antibiotic residues in farmed vannamei shrimp. Includes LOQs to set, a carton-level sampling plan, which labs can issue a COA in 3–5 days, and the exact PO clauses to use if a lot fails.
If you buy vannamei, you live with the nagging fear of nitrofurans or chloramphenicol popping up at the worst possible time. We have rejected our own lots and helped customers recover from detentions. Here is the protocol we use and recommend to buyers choosing between Indonesia, Vietnam, Thailand, and India.
How the risk really differs by origin
I am not a fan of sweeping generalizations, but residue risk clusters by how supply chains are organized, not just by country borders.
- Indonesia. Lower to moderate risk. Farms are typically small to medium but processors increasingly control pond selection and pre-harvest testing. Enforcement is improving and RASFF alerts are relatively low. You still need lot integrity and pre-shipment testing but the baseline is more forgiving than India in our experience.
- Thailand. Lowest risk. Consolidated industry, strong biosecurity, and robust internal testing. Price is higher and volumes are smaller, but antibiotic findings are rare.
- Vietnam. Moderate risk. Good processors and national labs, but multi-tier sourcing and occasional legacy practices still surface in RASFF for nitrofurans or malachite green. Manageable with strict lot control and testing.
- India. Higher risk. Big volumes and complex broker chains. There are excellent plants and many clean lots, but EU and FDA notifications for nitrofurans or CAP recur. If you buy Indian shrimp, do not skip pre-harvest and pre-shipment LC-MS/MS testing.
Takeaway. Country is a proxy. Your real protection is farm-level traceability and a lab-backed testing protocol tied to each lot.
What to test and which LOQs to require
EU and US rules differ, but two groups drive most detentions in shrimp.
- Nitrofurans. Test for AOZ, AMOZ, AHD, SEM as separate analytes. These are banned. EU treats them with effectively zero tolerance. We specify LOQ at or below 0.5 µg/kg per metabolite. Many labs can achieve 0.3–0.5.
- Chloramphenicol. Banned with zero tolerance. Set LOQ at or below 0.1 µg/kg.
What about other classes. Quinolones and sulfonamides have MRLs in the EU and are less common detention drivers in shrimp when farms are managed. If you ship to retailers or the EU, include them as a screen. Typical LOQs are 1–2 µg/kg for quinolones and 10 µg/kg for sulfonamides. Align with your target market’s MRLs.
We also see buyers include dyes like malachite green. That is not an antibiotic, but if your brand risk is zero-tolerance only, add MG and LMG with LOQ 0.5–1.0 µg/kg.
Practical panel for most PO’s. Nitrofurans (AOZ, AMOZ, AHD, SEM), chloramphenicol, quinolones, sulfonamides. If retail bound to EU, add MG/LMG.
How many cartons should you sample per lot
Here is the plan we use for frozen vannamei. It prioritizes detection while staying practical on cost and time.
- Define the lot correctly. One pond harvest or a homogeneous group of ponds harvested the same day and processed on the same line. Mixing ponds without tracking kills your protection.
- For up to 10 MT. Pull 10 cartons randomly from across the pallet stack map. Take 50 g increments from each carton to make two 500 g composites.
- For 10–20 MT. Pull 15 cartons. Same composite approach.
- For 20–25 MT. Pull 20 cartons. Same composite approach.
- Create duplicate composites. A-sample goes to the lab. B-sample stays frozen and sealed at the plant for re-test if needed.
- For peeled and value-added items. Keep a similar carton count but increase the number of small increments per carton. You want the composite to represent the entire production run, not a single work bench.
Most misses I have seen came from sampling only a handful of cartons from the top layer or testing a tiny grab of raw material then assuming the finished goods are clean. Test the finished product and spread your pull.
Can rapid kits replace lab analysis for nitrofurans or CAP
Short answer. No. Rapid ELISA or lateral flow kits are good for screening, especially pre-harvest. They are not accepted as confirmatory for export documentation. For your COA and any regulatory questions, you will need LC-MS/MS from an ISO 17025 lab with the right scope and LOQs. Use rapid kits to avoid obvious problems and save time, but do not ship without a lab COA.
Turnaround time and how to schedule testing before shipment
With a competent lab and pre-booked slots, you can get a COA in 3–5 working days for nitrofurans and CAP. Add shipping time if your samples travel to another city. Our typical schedule is:
- Day 0. Finish production. Pull samples. Courier to lab the same day with chain-of-custody.
- Day 1–2. Lab receives and logs in. LC-MS/MS starts.
- Day 3–5. COA issued. If you included a larger panel or MG/LMG, plan 5–7 days.
- Only load and seal once you have the COA. If you must load earlier, keep the container on hold locally until results clear.
Budget. As a rough guide in Asia, nitrofuran LC-MS/MS is often 150–250 USD per composite. CAP 60–120 USD. A broader antibiotic panel can add 70–150 USD. Rush fees of 25–50 percent are common. Costs vary by lab and country.
Which accredited labs can issue a COA in 3–5 days
We work with multiple ISO 17025 labs. Capacity shifts, but these networks consistently deliver in Asia.
- Global networks with seafood scope. SGS, Intertek, Eurofins, ALS, Bureau Veritas. Selected labs in Bangkok, Ho Chi Minh City, Jakarta, and Chennai regularly meet the LOQs above on LC-MS/MS.
- Strong local and regional options. Vimta Labs and Equinox in India. PT Saraswanti Indo Genetech and SUCOFINDO in Indonesia. There are others, but check accreditation scope and LOQ before you book.
Ask the lab for a method sheet showing LOQs for AOZ, AMOZ, AHD, SEM and CAP. Confirm they report non-detect at the LOQs you require. If you want our current short list or intros, Contact us on whatsapp.
What your PO should say if a lot fails
I have seen too many POs that say “zero tolerance” but do not define process. Add these clauses.
- Defined panel and LOQs. Name analytes and LOQs in the PO. Example. AOZ, AMOZ, AHD, SEM at LOQ ≤ 0.5 µg/kg each. CAP at LOQ ≤ 0.1 µg/kg.
- Lab and sample plan. Name acceptable ISO 17025 labs and the composite approach. Require A and B retains.
- Hold and release. Goods stay on hold until a clean COA is issued. No loading without release unless buyer approves in writing.
- Failure protocol. In case of detection above LOQ. Immediate lot quarantine. Confirmatory re-test on B-sample at a second pre-approved lab within 48 hours. If confirmed positive, supplier bears cost of replacement or refund, disposal, and any storage or demurrage. If negative and first result is not replicated, agree on a third-lab tie-breaker or accept the clean result.
- Traceability. Supplier must provide pond IDs, harvest date, and feed or chemical use records on request.
Clear language avoids emotional debates when time is burning.
How RASFF and FDA trends should guide sourcing
RASFF and FDA refusals are blunt instruments, but they do signal risk. Over the last year, India has seen more nitrofuran and CAP notifications in shrimp than Vietnam, while Indonesia and Thailand remain low. FDA Import Alerts on unapproved drugs hit firm by firm, but India and Vietnam show up more often in shrimp than Indonesia or Thailand.
What we do with this. We dial testing intensity to match. For India, we require pre-harvest screening plus finished goods LC-MS/MS on every lot. For Vietnam, finished goods LC-MS/MS on every lot is sufficient with trusted plants. For Indonesia, we still test every lot, but we are comfortable with smaller carton pulls if the plant has strong farm controls. For Thailand, we occasionally reduce panel breadth for non-retail customers, but retailers still get the full panel.
A quick note on Indonesian supply and how we work
We are Indonesia based and process farmed vannamei alongside wild-caught species. For vannamei, we select ponds with clean histories, run pre-harvest screens when needed, and issue lot-specific LC-MS/MS COAs before release. If you want Indonesian shrimp with this protocol baked in, start here. Frozen Shrimp (Black Tiger, Vannamei & Wild Caught). If you have a buyer spec you need us to match, Contact us on email.
Practical takeaways you can implement today
- Set LOQs. Nitrofurans at ≤ 0.5 µg/kg each. CAP at ≤ 0.1 µg/kg. Add MG/LMG for EU retail programs.
- Sample smart. 10, 15, or 20 cartons by lot size. Two 500 g composites. Finished product testing, not just raw.
- Choose a lab with the right scope. Pre-book slots to guarantee 3–5 day COAs.
- Write the PO for the worst day. Define failure protocol and cost allocation.
- Tune the protocol by origin. India needs the most intensity. Vietnam and Indonesia are manageable with strict LC-MS/MS on every lot. Thailand is lowest risk but rarely cheapest.
In our experience, buyers who lock in this system see fewer surprises and better pricing discussions because the rules are clear. And that is the point. Reduce uncertainty so you can spend time on quality and product mix, not firefighting residues.