Quitting smoking tied to less nerve damage in diabetes after six years, less kidney and eye damage after eight

People with type 2 diabetes who stop smoking are told it will help protect their heart. What it does for the kidneys, eyes and nerves, the organs that diabetes damages through its effect on small blood vessels, has been much less clear. A study of more than 200,000 Korean adults now suggests that a lower risk does appear, and that it takes years.

The study, published on October 2 in JAMA Network Open, found no lower risk in the first five years after quitting. A lower risk of the three complications taken together, and of nerve damage, came into view after six to seven years. For kidney damage and eye damage it appeared only after eight or more. The link was seen mainly in men, the researchers had no record of blood sugar control, and people who quit recently looked slightly worse for kidney disease than those who carried on.

Why smoking and small-vessel diabetes complications are studied

Long-term high blood sugar harms the body’s smallest blood vessels. The resulting conditions are called microvascular complications. Microvascular means relating to the smallest blood vessels. There are three main kinds: nephropathy, which is kidney damage; retinopathy, which is damage to the retina at the back of the eye; and neuropathy, which is nerve damage, most often in the feet and legs.

Together they are common. The paper describes them as affecting nearly half of patients with diabetes. It also sets out what is at stake with each: diabetic retinopathy is one of the major causes of vision loss worldwide, kidney damage from diabetes is a leading cause of kidney failure in many populations, and nerve damage is the most common reason for amputation of a foot or leg that does not follow an injury in most high-income countries.

Smoking is thought to add to the harm through systemic inflammation and endothelial dysfunction. The endothelium is the thin lining of blood vessels, and dysfunction means it is not working properly.

For the heart and large arteries, the case is long settled. A pooled analysis from 2015 covered people with diabetes specifically, and a total of 89 cohort studies were included. Smokers with diabetes had a 55% higher risk of death than those who did not smoke, and the authors concluded that smoking cessation is associated with reduced risks in comparison with current smoking.

For the small vessels the picture has been patchier. A scoping review published in 2024 searched for every human study of quitting and diabetes complications. It identified 1,023 studies, and only 26 met the criteria for eligibility. Its authors found that regarding microvascular complications, the strongest evidence for the beneficial effects of smoking cessation is observed in diabetic nephropathy. By contrast, the relationship between smoking cessation and retinopathy, neuropathy and other complications had been little studied, they wrote. None of that work could say how long a person has to stay quit before risk begins to fall.

How the study tracked quitting in people with diabetes

The research was led by Soonsu Shin, with Jae-Hong Ryoo as senior author, and was approved by the ethics board of Kyung Hee University Hospital in Korea. It drew on the National Health Insurance Service database, which the authors describe as a single-payer system covering nearly the entire Korean population. The database holds medical claims and also the results of health checkups offered every two years to adults aged 40 and over, where people are asked about smoking.

The study took adults first diagnosed with type 2 diabetes between 2003 and 2013. To be included, a person needed at least four checkups, no more than about three years apart, with the last one falling between 2009 and 2013. People who already had one of the three complications were excluded.

Smoking status had to be consistent across every checkup. People classed as never smokers said so each time. Continuous smokers reported smoking each time. Quitters were those who reported current smoking at their initial health checkup and then reported having stopped at every later one, with no relapse. People who took up smoking during the period, or whose answers changed more than once, were left out.

From 1,599,870 people with type 2 diabetes who had a checkup in those years, the criteria left 209,698. Of these, 162,861 had never smoked, 34,989 smoked throughout and 11,848 were sustained quitters. The average age was 57.2 years and 61.0% were women. Those who had never smoked were older, at an average of 58.8 years, than quitters (53.0) or continuous smokers (51.1).

For each quitter the team estimated a quit date. In the paper’s description, the date of smoking cessation was approximated as the midpoint between the health checkup at which cessation was first reported and the preceding checkup. Time since quitting was then sorted into five bands: under 2 years, 2 to 3, 4 to 5, 6 to 7, and 8 or more.

Everyone was followed from their last checkup until a first diagnosis of kidney, eye or nerve damage appeared in their insurance records, or until death or the end of 2023. The comparisons were adjusted for age, sex, body mass index, blood pressure, cholesterol, a liver enzyme, alcohol, physical activity, high blood pressure, lifetime amount smoked and years since the diabetes diagnosis.

What the study found about years since quitting smoking

The raw rates set the scene. Among continuous smokers there were 301.9 new cases of any of the three complications for every 10,000 person-years. A person-year is one person followed for one year. Among never smokers the rate was 287.6. The highest rate of all was among people who had quit less than two years earlier, at 312.8. It then fell with each longer band, reaching 237.6 among those who had quit eight or more years before.

After adjustment, and with continuous smokers as the comparison:

• Under 6 years. No band showed a lower risk of the combined outcome.
• 6 to 7 years. Risk of any of the three complications was 12% lower, with a range from 1% to 22%. A range here is the margin of uncertainty around an estimate. Risk of nerve damage was 24% lower, with a range from 11% to 36%. Kidney damage and eye damage were not lower in this band.
• 8 or more years. Risk of any complication was 19% lower, with a range from 5% to 30%. Kidney damage was 21% lower (6% to 36%), nerve damage 20% lower (2% to 34%) and eye damage 41% lower (3% to 64%).

The eye result has by far the widest range. The authors say this estimate was based on a small number of events and should be interpreted with caution.

Treated as a sliding scale and averaged across all quitters, each additional year without smoking went with about 1% lower risk of the combined outcome, with a range from 1% to 2%.

Two further results complicate the story. First, the penalty attached to smoking itself was small in these data. Continuous smokers had a 3% higher risk of any complication than never smokers, with a range from 1% to 5%, and a 9% higher risk of nerve damage. Second, the most recent quitters looked worse, not better. Those who had stopped less than two years earlier had a 12% higher risk of kidney damage than people still smoking, with a range from 3% to 23%.

The paper’s conclusion is that any reduction in risk required sustained, long-term cessation rather than appearing soon after quitting.

Why the smoking results were clear only in men

When men and women were analyzed separately, the pattern held in men. Among men who had quit for eight or more years, risk of any complication was 18% lower than in men who kept smoking, with a range from 5% to 30%.

In women it could not be assessed. The paper says female-specific analyses were limited by small sample size. Women made up most of the study, but the great majority of them had never smoked, which left few female quitters and smokers to compare. The authors draw the cautious conclusion that the results may not be generalizable to female patients.

How this fits with other research on quitting smoking in diabetes

The same research group had looked at the question before, in men only. That 2023 study followed 26,673 men with diabetes who had health checkups in both 2003-2004 and 2009. Its authors concluded that lasting and heavy smoking increases the risk of microvascular complications, including nephropathy and neuropathy. It also found a higher risk among men who had moved from smoking to not smoking between the two checkups, the same early bump seen in the new data.

Why would recent quitters do worse? One explanation is that illness prompts quitting. Another has to do with blood sugar. A UK study of 10,692 smokers with type 2 diabetes in primary care records found that HbA1c, a blood test reflecting average blood sugar over about three months, rose after people quit. The authors reported a deterioration in control that lasts for 3 years and is unrelated to weight gain. They warned that, across a population, this temporary rise could increase microvascular complications. A rise in blood sugar in the first years after quitting would fit a benefit that takes time to show.

Shorter-term studies have pointed in a hopeful direction for the kidneys. In Greece, researchers followed 193 smokers newly diagnosed with type 2 diabetes who already had small amounts of protein in their urine, an early sign of kidney damage. A total of 120 (62.2%) subjects quit smoking over the year. Of the treatments and lifestyle changes the authors examined, quitting was the one most strongly and independently tied to the reduction of microalbuminuria. Microalbuminuria is the term for those small amounts of protein in urine.

On the heart, the benefit shows sooner. An analysis of people with type 2 diabetes drawn from two prospective cohorts in the USA found that those who had quit two to six years earlier had a 17% lower risk of cardiovascular disease than continuing smokers, and long-term quitters a 28% lower risk.

Guidance does not hinge on the small-vessel question. This year a multidisciplinary panel of 35 experts from 13 countries issued consensus recommendations saying that adults with type 2 diabetes who smoke require integrated, high-intensity cessation support throughout the quitting process. The new paper notes that about one in five people with type 2 diabetes worldwide still smokes.

What the study of smoking and diabetes complications cannot show

Cause and effect. This was an observational study. It compared people who made different choices and cannot prove that quitting produced the lower risk.

No blood sugar data. The database did not hold HbA1c results. The authors adjusted for how long people had had diabetes as a stand-in, but concede that residual confounding by glycemic control cannot be fully excluded. Confounding means that some other difference between groups accounts for part of a result, and glycemic control means control of blood sugar. Blood sugar is the main driver of these complications.

Quitting because of illness. In the paper’s words, the possibility of sick quitter bias cannot be excluded. Some participants may have stopped smoking because of worsening health status or early manifestations of diabetic complications, which would make recent quitters look worse than they otherwise would. The raised kidney risk in the first two years fits that explanation.

Healthier in other ways. The opposite bias may flatter long-term quitters. In this study quitters were more likely to exercise regularly and less likely to drink frequently than continuous smokers, and habits that were not measured may differ too. One result points the same way. People who had quit for eight or more years had a 19% lower risk of any complication than people who had never smoked, with a range from 6% to 30%, which is hard to explain by quitting alone.

A selected group. Only people with at least four checkups and a perfectly consistent smoking history were kept, about one in eight of those screened. They may be more health-conscious than most people with diabetes.

Smoking was self-reported and then frozen. Status came from questionnaires, and the paper states that smoking status and cessation duration were fixed at the index date and were not updated during follow-up. Anyone who quit or relapsed later was still counted in their original group.

Diagnoses from insurance codes. Complications were identified from billing records, which depend on a person being tested and the result being coded.

Women, and other countries. The study could not assess women separately with any confidence, and all participants were Korean.

Whether and how a particular person with diabetes is helped to stop smoking is assessed by their doctor.

In these records the lower risk of small-vessel damage showed up only in people who had stayed off cigarettes for most of a decade.