Anita Albrecht said she was told during a routine appointment she was ‘eating too much’ and needed to lose weight.
The 39-year-old, who works as a personal trainer, said her body mass index came out at 29 – four points over the healthy range and one short of obese.
But Ms Albrecht, who competes against some of the world’s leading bodybuilders, said the measure was distorted by her muscle-bound physique.
She said: ‘She insulted me by making assumptions about my lifestyle.
‘The information the nurse has given me is actually dangerous. A 1,000 calorie-a-day diet is only for people who are severely obese who are not active.
‘They should only be on that for a maximum of 12 weeks – generally as a precursor to surgery.’
BMI is worked out by dividing weight in kilograms by height in metres squared. But critics say it is flawed because it does not distinguish between fat and muscle, which is heavier.
Miss Albrecht, who is 1.5m (4ft 11in) tall and weighs about 66kg (10st), saw the nurse during an appointment about contraception at a family planning clinic in Harold Hill, east London.
She was told she needed to exercise more, eat less and to cut alcohol and fruit juice from her diet.
‘She put me on scales and clearly I’m a lot heavier than other women because of my height and I’m a bodybuilder,’ said Miss Albrecht.
‘For nine months of the year I don’t even drink as I am a competitive athlete. I felt insulted, was made to feel as though I was overweight, over eating and I felt a knock in my confidence.
‘When I tried to explain to her about body composition she wasn’t interested at all.’
NHS England declined to comment because of patient confidentiality. But its website concedes that athletes, such as rugby players, can be wrongly classed as overweight using BMI.
Body Mass Index is calculated by dividing a person’s weight in kilograms by their height in metres squared.
The measurement was devised by Belgian mathematician Adolphe Quetelet while studying astronomy in 19th century France.
The student was inspired by stargazers who used probability to make their measurements as accurate as possible and decided to apply the same techniques to human features.
However, it wasn’t until the 1970s that researchers re-visited and supported his ‘Quetelet Index’ formula.