In today's health-conscious environment, most people are aware that lifestyle can have a significant impact on causing or preventing many diseases, and the patients in your office are no exception. Curiosity about this is particularly evident with patients who are newly diagnosed with primary open-angle glaucoma. Once you explain to them what the problem is and what you're going to do to treat it, they often ask, "What else can I do to slow or halt the progression of the disease besides taking the medications you're prescribing?"

It's important to choose your response carefully, because any advice you give the patient needs to be backed up by good clinical data. At present, the reality is that the evidence is mixed in terms of the benefits of modifying any lifestyle factor as a means to prevent glaucoma. That doesn't mean that correlations to lifestyle don't exist; it just means we haven't found any definitive ones yet.


What Do We Know?

Given the complexity of POAG, which probably comprises multiple diseases, it's likely that the patient's genetic background is a key to how the environment affects the presence or progression of glaucoma. Consider a hypothetical male patient in your exam lane with a new diagnosis of POAG: The risk of his 65-year-old next-door neighbor also having POAG is about 2 percent, but the risk of his sibling developing POAG is closer to 6 percent.

It appears likely that each patient has a unique genetic background that contributes to the development or non-development of glaucoma within that person's environmental context. If that's true, then whatever lifestyle modification advice we eventually can offer with confidence won't be "one size fits all"; it may have to be tailor-made to fit each patient's genetic background. (However, it's worth noting that the risk for his identical twin would only be about 10 percent, much lower than you would expect if genetic factors alone contributed to the disease.1)

What about smoking, exercise and diet? We've done extensive research on potential associations to these factors at Harvard; Jae H. Kang, ScD, of the Channing Lab at Brigham and Women's Hospital, has been lead author on multiple papers discussing smoking, alcohol intake, dietary fat and antioxidants and their relation to incident primary open-angle glaucoma.2-5 Again, no conclusive evidence has been found that altering these will affect the course of glaucoma for any given individual.

It's important to note that smoking is deleterious to your health; it increases your risk of lung cancer, macular degeneration and other diseases. Clearly, smoking cessation saves lives, so I encourage all of my patients who smoke to quit. But I try not to mislead patients into thinking that quitting will help their glaucoma, because I don't want to leave patients with the impression that if they stop smoking they can then skip taking their glaucoma medication every day.

The same is true for exercise. Cardiovascular benefits to exercise have been documented time and time again, and we do know that exercise lowers intraocular pressure—but we don't know whether that necessarily translates into a reduced risk of POAG. One study looked at the relationship between vigorous physical activity and risk of glaucoma; it noted a reduced risk of glaucoma among the most physically fit male runners.6 However, in that study the definition of glaucoma is unclear, so we don't know what kind of glaucoma they were measuring. Of course, all of these data do suggest that there is some sort of correlation, so we should encourage patients to exercise. But at the same time we need to get across the message that patients will still need to use their glaucoma medications.

In terms of advising your patients, I believe the best answer is to recommend eating the diet that their internists recommend for overall good health. Right now, we simply don't know what might constitute a glaucoma-protective diet.


Exploring the Weight Connection

Among other things, diet and exercise affect a person's body-mass index (defined as an individual's body weight in kilograms divided by the square of his height in meters). Generally speaking, BMI is correlated with adiposity in middle-aged and older adults. So, in the hope of finding novel factors associated with POAG, we recently conducted a study of possible correlations between anthropomorphic measures such as BMI and glaucoma.7

We undertook the study because of suggestive data we uncovered while doing a 2006 study of the relationship between type II diabetes and POAG.8 That study found a 1.8-fold increased risk of POAG among subjects with type II diabetes mellitus. We controlled for BMI in that study because BMI is strongly related to type II diabetes; if your BMI is 30 kg/meter2 (which would make you obese by World Health Organization standards), then your risk of getting diabetes is 39-fold greater than a person with a BMI of 22 kg/meter2 (typical for a lean person).

The apparently inverse relationship between BMI and POAG that we noted was especially interesting given other data in the literature that suggests that heavier people tend to have higher eye pressures. Also, heavier people who develop type II diabetes mellitus tend to have higher blood sugars and such people also tend to have higher IOP. So, we decided to do a formal analysis of the relationship between BMI and POAG.

To conduct this study we used data from 78,777 female participants in the Nurses' Health Study, and 41,352 male participants in the Health Professional Follow-up Study. The subjects comprise a prospective, longitudinal cohort; every two years the subjects were asked about their lifestyle, what they ate and drank, how much they exercised, and what their weight and height were (along with other body shape features). They were also asked whether they received a diagnosis of glaucoma.


What the Data Showed

Several interesting correlations turned up—and failed to turn up—in our data. The primary result was that we found no positive relationship between BMI and POAG. In other words, in this population, heavier people were not more likely to get POAG, despite the fact that heavier people are statistically more likely to have diabetes and elevated IOP.

We also looked at weight change; we asked people what they weighed when they were in early and middle adulthood, and then every two years from 1986 on. We didn't find weight change to be related to glaucoma at all. That means that although telling your patient to lose weight may be a good idea for other reasons, it doesn't appear that losing weight affects a person's risk of developing glaucoma.

We did, however, find an inverse correlation between BMI and the likelihood of developing normal-tension glaucoma—i.e., glaucoma with an IOP less than 21 mmHg—among women. Thin women were significantly more likely to develop NTG than heavier women. However, no such correlation appeared for the men. (See table.)

Furthermore, this correlation was also true for BMI earlier in life (as recalled by participants). That's interesting, because although we think of glaucoma as a disease that occurs in the later decades of life, optic nerve changes—or biochemical changes that later precipitate the disease—may be starting at a much younger age. These findings were secondary findings in our study, but they were strong enough and intriguing enough to deserve some attention and speculation about possible explanations.

So: How might a low BMI contribute to the development of POAG at normal IOP in women? A thin woman has a lower adiposity reserve than a woman with a higher BMI; adipose tissue may act like an endocrine organ, secreting a protective factor that keeps retinal ganglion cells alive.

One possible protective factor might be estrogen; there is emerging evidence that estrogen is a neuroprotective agent. (If you give estrogen to animals with elevated IOP, you find less ganglion cell loss than in animals not receiving estrogen.9) And, we know that a heavier woman, even postmenopausal, generates more estrogen than a lean woman. Given the multifactorial nature of this condition, perhaps a little more estrogen may help to prevent the development of POAG optic neuropathy among women who are predisposed to develop open-angle glaucoma despite normal IOP.

Unfortunately, this speculation still gives rise to the question: Why doesn't this correlation appear in men? All retinal ganglion cells have estrogen receptors on them, and this is not specific to women—men also have them. Furthermore, why is there no apparent relation between BMI and high-tension POAG in women? At this point, we have no answers for these questions.

It's worth noting that the subjects in our database are 99 percent European-derived Caucasians, so these findings may not apply to people of African descent. In fact, Cristina Leske, MD, MPH, found an inverse relationship between BMI and POAG in the Barbados eye study among people of African/Caribbean descent—but her group found it in men more than women, the opposite of our findings.10 So, the biology involved could be significantly different in this population.

The only correlation our data showed that didn't change with gender or POAG stratified by IOP was that a greater hip circumference was associated with a reduced risk of POAG. This result wasn't strong, as the p for trend was 0.11. Nevertheless, those subjects with the largest hip circumference were 27 percent less likely to develop POAG than those with the smallest hip circumference. (We didn't find this correlation with waist circumference, but hip circumference is more of a reflection of a person's muscle and bone mass; waist circumference is generally a reflection of adipose tissue.)

What could account for this association? Our speculation is that something secreted from muscle and bone may be protective in the development of this disease. In any case, this data needs to be confirmed with further clinical studies. (It would certainly be premature to be less concerned about a glaucoma suspect because he or she has a large hip circumference.)


What Should We Say?

Given these data, how should we answer when patients ask about non-medical changes in lifestyle they might make? We can mention that there is some evidence that exercise may be beneficial, though it's far from conclusive. I think the main message that needs to be communicated to these patients is that adopting a healthy lifestyle—i.e., not smoking, getting exercise and eating sensibly—is likely to be beneficial, including lowering the risk of many other eye problems (including macular degeneration). At present we simply don't know whether the benefits of these changes include lowering the risk of POAG.

Most important, we need to make sure that patients understand that if they want to preserve their eyesight, there's no substitute for following their prescribed treatment.

Dr. Pasquale is director of the glaucoma service and associate director of telemedicine at the Massachusetts Eye and Ear Infirmary; clinical assistant scientist at Schepens Eye Research Institute; and associate professor of ophthalmology at Harvard Medical School.


1. Teikari M. Genetic factors in open-angle (simple and capsular) glaucoma. A population-based twin study. Acta Ophthalmol 1987;65:6:715-20.

2. Kang JH, Pasquale LR, Willett W, Rosner B, Egan KM, Faberowski N, Hankinson SE. Antioxidant intake and primary open-angle glaucoma: a prospective study. Am J Epidemiol 2003;158:4:337-46.

3. Kang JH, Pasquale LR, Rosner B, Willet W, Egan KM, Faberowski N, Hankinson SE. Prospective study of cigarette smoking and the risk of primary open-angle glaucoma. Arch Ophthalmol 2003;121:1762-68.

4. Kang JH, Pasquale LR, Willett WC, Rosner BA, Egan KM, Faberowski N, Hankinson SE. Dietary fat consumption and primary open-angle glaucoma. Am J Clin Nutr 2004;79:5:755-64.

5. Kang JH, Willett WC, Rosner BA, Hankinson SE, Pasquale LR. Prospective study of alcohol consumption and the risk of primary open-angle glaucoma. Ophthalmic Epidemiology. 2007;14:3:141-7.;

6. Williams PT. Relationship of incident glaucoma versus physical activity and fitness in male runners. Med Sci Sports Exerc 2009;41:8:1566-72.

7. Pasquale LR, Willett WC, Rosner, BA, Kang JH. Anthropometric Measures and their Relation to Incident Primary Open-Angle Glaucoma. Ophthalmology 2010. [Epub ahead of print].

8. Pasquale LR, Kang JH, Manson JE, Willett WC, Rosner BA, Hankinson SE. Prospective study of type 2 diabetes mellitus and risk of primary open-angle glaucoma in women. Ophthalmology. 2006;113:7:1081-86.

9. Russo R, Cavaliere F, Watanabe C, Nucci C, Bagetta G, Corasaniti MT, Sakurada S, Morrone LA. 17Beta-estradiol prevents retinal ganglion cell loss induced by acute rise of intraocular pressure in rat. Prog Brain Res 2008;173:583-90.

10. Leske MC, Connell AM, Wu SY, et al, Barbados Eye Study  Group. Risk factors for open-angle glaucoma: The Barbados Eye Study. Arch Ophthalmol 1995;113:918 –24.