CLINICAL REFERENCE FOR HEALTHCARE PROFESSIONALS
Two diseases on one plate, pulling in opposite directions. This chapter sets out what the current guidelines actually say, applies it to Indian food with values per katori and per roti, and names the numbers still being taught here that are in no guideline at all.
Who this is written for. Dieticians, physicians and nursing staff. It is clinical education, it assumes clinical training, and it is not a patient handout.
It is not advice about any medicine. Medicines are named only to explain how they change what a dietician must do at the counselling table. Nothing here is a recommendation, offer or advertisement of any medicine, and no brand or trade name of any medicine appears anywhere in this chapter. Prescribing decisions belong to the treating doctor.
If you are a patient who has been told you have kidney disease and diabetes, do not use this chapter to change your diet. Bring it to your own doctor or dietician, or see the dietetics department.
Start here. Every difficult conversation in this clinic comes from one of the six conflicts below, and a counsellor who has not named them will keep giving advice that cancels itself out.
A diabetes diet and a renal diet were designed by different people for different organs. Put them on one plate and they argue. The job is not to average them. The job is to know which one wins in this patient, at this eGFR, on these drugs, this month.
| The diabetes diet wants | The renal diet has historically wanted | How we resolve it now |
|---|---|---|
| High fibre: whole grains, dals, fruit, vegetables | Restrict potassium, which means restricting exactly those foods | Restrict bioavailable potassium (processed food, additives, salt substitutes) rather than produce. Leach and portion produce only if the patient is actually hyperkalaemic. |
| Weight loss in obesity | Protect against protein-energy wasting | Intentional weight loss is a goal at eGFR ≥30. Below that, and on dialysis, the goal shifts to preserving lean mass. Transplant listing is the exception. |
| Low glycaemic load staples | Milled rice and maida, which are the lowest potassium and phosphate staples we have | Control the glycaemic load with portion size, protein and fat pairing, and sequence, not by swapping to millets in a hyperkalaemic patient. Millets carry 3 to 4 times the potassium of milled rice. |
| Generous fluid intake | Fluid restriction on dialysis and in oedema | Fluid follows urine output, oedema and interdialytic weight gain, not thirst advice. Hypoglycaemia is treated with glucose tablets, not fruit juice. |
| Carbohydrate restriction, sometimes ketogenic | No position, historically | Hard stop. Almost every diabetic CKD patient is now on an SGLT2 inhibitor. Ketogenic eating plus an SGLT2 inhibitor is a euglycaemic ketoacidosis setup, and cutting carbohydrate pushes protein up against a 0.8 g/kg ceiling. |
| Milk, curd and paneer as convenient protein | Limit phosphate | Dairy phosphorus has no phytate and is 40 to 60 percent absorbed. Paneer is superb for potassium and poor for phosphate. Pick the food for the abnormality you actually have. |
KDIGO 2024 says plainly that RAS inhibitors, SGLT2 inhibitors and GLP-1 receptor agonists deliver better kidney outcomes than dietary protein restriction does. That is not a reason to stop counselling. It changes what counselling is for.
The diet's first job in diabetic CKD is to keep the patient on the drugs that have grade 1A evidence. Bring potassium down so finerenone and the ACE inhibitor can continue. Bring sodium down so the blood pressure target is reachable. Keep carbohydrate in so the SGLT2 inhibitor is safe. Protect protein and energy so the GLP-1 agonist does not tip the patient into wasting. Glycaemia and blood pressure are the second job. Protein restriction is now adjunctive, not the centrepiece.
If a staff member remembers nothing else from this chapter, these ten sentences should survive.
Neither KDOQI 2020 nor KDIGO 2024 gives a milligram ceiling for potassium or for phosphorus. Both say the same thing: adjust intake to keep the serum level in range. A patient with a normal potassium does not need a potassium restriction, and giving one causes harm.
Stack potassium plus phosphate plus sodium plus protein restriction and you have deleted fruit, vegetables, dals, whole grains and dairy from an Indian diet. What is left is white rice and oil. Protein-energy wasting affects 11 to 54 percent of non-dialysis CKD G3 to G5 and most dialysis cohorts fall between 28 and 54 percent, and it kills faster than the electrolyte you were chasing.
It is one of the very few dietary interventions in CKD supported by high-certainty Cochrane evidence: a realistic reduction of about 4 g salt per day lowers blood pressure by 6.9 over 3.9 mmHg and cuts albuminuria by 36 percent. Nothing else on this page has evidence that good.
Potassium from fruit and vegetables is roughly 50 to 60 percent absorbed. From animal foods, about 80 percent. From additives and potassium-based salt substitutes, close to 100 percent, with no food matrix and no phytate. A patient using a 66 percent potassium-chloride substitute such as LoSalt is eating more absorbable potassium than a patient eating three bananas.
Phytate-bound plant phosphorus is 20 to 40 percent absorbed. Animal phosphorus 40 to 60 percent. Inorganic additive phosphate 90 to 100 percent. Banning dal and atta while the patient keeps drinking cola and eating processed cheese is the commonest error in Indian renal counselling.
Write the reference weight on the chart. ESPEN is the only body that defines it, and it is explicit: actual body weight should not be used for the protein prescription. Use pre-illness or usual body weight, and for obesity the weight corresponding to a BMI of 25.
Acute illness, sepsis, wasting, a falling albumin, unintentional weight loss: every low-protein statement in every guideline is switched off. KDIGO says it as a practice point in plain words. Protein restriction is contraindicated here, not merely relaxed.
Insulin clearance is 80 percent renal. As eGFR falls, insulin requirement falls, and hypoglycaemia risk roughly doubles. Meanwhile HbA1c itself becomes unreliable and on average underestimates true glucose. Congratulating a dialysis patient on a 6.6 percent HbA1c is how we miss both hyperglycaemia and asymptomatic hypoglycaemia in the same person.
Sodium bicarbonate 650 mg three times a day is 530 mg of sodium. Sodium zirconium cyclosilicate 10 g is 800 mg of sodium. Phosphate binders only work if they are timed to the meal, not the clock. Half of what you are trying to fix is being prescribed.
No dietary pattern, Mediterranean, DASH-modified, plant-dominant or low-protein, has hard outcome evidence in diabetic kidney disease specifically. Since the evidence does not choose the pattern for you, choose the one the family will actually cook. In this clinic that means rice, dal, roti, gourds and curry leaves, re-engineered, not replaced.