On this page
- The word on the report: borderline
- The borderline sugar level range — the full table
- Why one lab calls 105 normal and another borderline
- Is borderline sugar dangerous?
- Borderline diabetes symptoms — and why most people have none
- What happens next is not fixed
- What to do in the next four weeks
- Related
- Frequently asked questions
The word on the report: borderline
Every week a report crosses this desk with the fasting sugar circled in blue ink and one word beside it: borderline. By the time the patient sits down, they have usually already decided what it means, and the decision goes one of two ways — either the report is a sentence, or it is a pass mark, filed away with relief because nothing was actually wrong. Both readings are wrong, and the second one is the expensive one.
Borderline sugar is the stage medicine calls prediabetes: glucose regulation that has begun to slip but has not yet crossed the diagnostic line. Diabetes itself is sugar noi in Tamil and prameham in Malayalam; the stage before it has no inherited word in either language, which is exactly why the English “borderline” now travels through every household in this district — on WhatsApp, at weddings, across the counter at the lab. A word that common deserves a better definition than the one it usually gets.
The borderline sugar level range — the full table
Three tests define the band, and any one of them can put you inside it. All values are venous laboratory samples in mg/dL; HbA1c is reported as a percentage.
| Test | Normal | Borderline (prediabetes) | Diabetes |
|---|---|---|---|
| Fasting (8–12 hrs, water allowed) | below 100 | 100–125 | 126 or above |
| 2 hours after meal / 75 g glucose (OGTT) | below 140 | 140–199 | 200 or above |
| HbA1c (3-month average) | below 5.7% | 5.7–6.4% | 6.5% or above |
You do not need all three to agree. A fasting sugar of 96 alongside an HbA1c of 6.0 is still borderline sugar, and so is a comfortable fasting value with a two-hour reading of 165 — which is why the glucose tolerance test, the most informative and the most skipped of the three, catches people the fasting test quietly clears. Note also what the table does not contain: a random sugar taken an hour after lunch at a camp is a screening prompt, not a diagnosis. If you want the full diagnostic picture including treatment targets, the blood sugar levels chart sets out every band, and the HbA1c test explained covers what that percentage is actually measuring.
Why one lab calls 105 normal and another borderline
This is the question that brings people back with two reports and a grievance, and it has an honest answer: there are two respectable cutoffs in the world, and Indian laboratories print whichever one their reference range was built on. The American Diabetes Association calls fasting glucose abnormal from 100 mg/dL. The World Health Organization’s impaired-fasting-glucose band starts at 110 mg/dL. So a value of 105 is genuinely “borderline” on one report and genuinely “within normal limits” on another — not an error, not a scam, and not something the technician can be argued out of.
Both cutoffs are defensible. The lower one catches more people early and labels many who would never have progressed; the higher one labels fewer and misses some. What neither cutoff can do is tell you about you, and that is where the argument should end. Three things settle it in practice: an HbA1c, which does not care which fasting threshold your lab prefers; a repeat test under proper conditions at the same laboratory; and the direction of travel across those reports — a 105 that was 92 two years ago is telling a different story from a 105 that was 108 last year. Trend beats threshold. Bring both reports to a consultation rather than choosing the kinder one.
Is borderline sugar dangerous?
Not today, and not in the way patients fear — nobody is admitted for a fasting sugar of 112, and there is no emergency hiding in that number. The danger is a trajectory, not an event. In a Kerala cohort study, about six in ten people with borderline fasting sugar had developed diabetes within ten years. An Indian trial found that among adults with borderline glucose who received ordinary advice alone, more than half had crossed into diabetes within three years. Nationally, the ICMR-INDIAB survey published in 2023 counted 136 million Indians with prediabetes alongside 101 million with diabetes — a queue longer than the diagnosis it feeds.
Two further points matter for Indian bodies. Risk to the heart and blood vessels rises through the borderline band rather than switching on at 126 — the diagnostic line is an administrative convenience, not a biological cliff. And South Asians reach the diabetes risk of a heavyset European at a body mass index of about 22, which is why a slim uncle with a soft midsection and a borderline report is an entirely ordinary patient here, not a paradox. The reassurance “it’s only borderline” is the one sentence that reliably converts this stage into the next.
Borderline diabetes symptoms — and why most people have none
The honest answer first: most people with borderline sugar feel completely well, which is precisely why 136 million of them are walking around undiagnosed. The classic triad taught in schools — excessive thirst, frequent urination, constant hunger — belongs mostly to established, higher sugars. Waiting for it before testing means waiting for the stage you were trying to avoid.
What does sometimes show up is quieter and easily blamed on age or work: a dark, velvety discolouration on the neck, underarms or knuckles that no amount of scrubbing removes; skin tags around the neck; heaviness and sleepiness an hour after a large rice meal; cuts and scrapes taking longer to close; recurrent boils or fungal itch in skin folds; a waist expanding while the weighing scale stays flat. None of these prove anything on their own, and their absence proves even less. If you have a family history, a waist that has outgrown two belts, or a past pregnancy with high sugar, the symptom question is the wrong question — the test is the answer. The 60-second diabetes risk score sorts out whether you belong in the queue, before any needle.
What happens next is not fixed
The same trial that measured how fast untreated borderline glucose progresses also measured what happens when people act on it: sustained lifestyle change — modest weight loss, daily movement, a steadier plate — cut new diabetes by nearly a third among Asian Indian participants. That is a result obtained on bodies like ours, in a Chennai programme, not imported from a Finnish cohort and hoped upon.
The working numbers are unglamorous and consistent across prevention trials: about seven percent of body weight, which is five kilos for a seventy-kilo man; thirty minutes of brisk walking on most days, which can be split into pieces after meals; and a plate where rice is measured rather than heaped. For Indian bodies the measuring tape outranks the weighing scale — the waist is where our risk is stored. None of this requires a package, a powder or a plant, and anyone selling a normal report in twenty-one days is selling the calendar rather than the biology. The full method, including what “reversed” means clinically and why it is maintained rather than won, is set out in how to reverse prediabetes.
What to do in the next four weeks
Four steps, in order. Repeat the test properly — a fasting sample after eight to twelve hours without food (water is fine, tea with sugar is not), with an HbA1c drawn alongside it, at the same laboratory you will keep using. Learn what your particular number means: every decimal between 5.7 and 6.4 carries a slightly different message, laid out value by value on the HbA1c 5.7 to 6.4 page. Change the plate before you change anything else, using food that already exists in a Kanyakumari kitchen rather than imported substitutes — the borderline sugar food list covers rice, kappa, idli, chips and payasam in the portions people here actually eat. And book the review at three months, because HbA1c cannot show your effort any sooner than that.
If you have never been tested at all and arrived here on a suspicion — a family history, a dark neck, a father with sugar noi — stop estimating. One blood test settles it. Book a test, get the fasting value and the HbA1c together, and come back to this page with numbers instead of worry. Borderline is not bad news. It is the last stage at which the news is still yours to change.
Related
Frequently asked questions
Borderline. A fasting value of 110 mg/dL sits inside the prediabetes band of 100–125, and it is nowhere near diabetes, which requires 126 or above confirmed on a second occasion. It is also the exact value where laboratories disagree with each other, so repeat it properly after eight to twelve hours of fasting and add an HbA1c.
Both, and together they place you in the borderline band — one abnormal test is enough. Fasting sugar is a photograph of one morning; HbA1c averages roughly three months of blood and is harder to flatter with a careful night before. Do not pick the friendlier number. Treat the abnormal one as the signal and repeat in three months.
For most people, no. Lifestyle change is the first treatment and it outperformed medication in several prevention trials. Some patients — younger, heavier, with readings at the top of the band or a history of gestational diabetes — may be prescribed medication after assessment. That is a decision made in a consultation with your numbers in hand, never one to start or stop on your own.
Every three to six months while you are actively working on it, because HbA1c needs about three months to show any change. Once the numbers come back to normal, once a year, indefinitely. Use the same laboratory and the same conditions each time — the trend across three reports says far more than any single value.
No. Diabetes is a diagnosis that stays with you and is managed toward remission at best; borderline sugar is a stage that can genuinely return to normal regulation and stop being a diagnosis at all. The distinction is worth defending, because it is the difference between managing a condition for life and closing the door before it opens.