Methods

The study was conducted at Kaiser Permanente Southern, a group-model health maintenance organization that provides care for 3.2 million residents of Southern California with over 7000 physicians working in 12 medical centers. The study is a case-control study to determine whether incident cases of cataract extraction are less likely to be users of statins. Eligible patients were 50 years or older and had been enrolled in Kaiser Permanente Southern California for at least 5 years in 2009. Cases were any patient who had cataract extraction, identified with the CPT (current procedural terminology) code 66982, 66983, or 66984, in the year 2009 and who did not have this procedure in the past. Controls were patients who had an eye examination in 2009 but who had not had cataract extraction and did not have a diagnosis of cataract anywhere in their chart (366.x).

The primary source of data to assess treatment with statins is our electronic database containing outpatient prescriptions filled in Kaiser Permanente pharmacies, called the Pharmacy Information Management System. The Pharmacy Information Management System has been used in Kaiser Permanente Southern California since 1991, with complete data available from January 1992. This system actually consists of several databases, with records linked between databases using medical record numbers and other identifiers. Information available from the Pharmacy Information Management System includes details on known drug allergies as well as the following data for each prescription dispensed: date, place, and person filling the order; the type of medication dispensed, identified by Generic Product Identification code; number of units dispensed; dosage; route of administration; ingredients (if medication mixed by pharmacist); and billing information. About 93% of Kaiser Permanente Southern California Health Plan members have prescription drug coverage as a benefit. It is estimated that more than 97% of all prescriptions written by Kaiser Permanente physicians are filled at Kaiser Permanente Southern California pharmacies and therefore would be included in the Pharmacy Information Management System.

Statin use was defined as use of atorvastin, ezetimibe-simvastatin, lovastatin, pravastatin, and simvastatin. In analyses involving other lipid-lowering agents, cholestyramine, colestipol, ezetimibe, fenofibrate, and gemfibrozil were included. Drug use was defined as use prior to case determination. Recent use was defined as a filled prescription of statins in the year prior to the year of diagnosis and recent longer-term use, a filled prescription in each of the 5 years prior to the diagnosis. Coronary artery disease was defined as: 1) any hospitalization with a diagnosis of myocardial infarction or coronary artery disease; 2) procedure of coronary artery angioplasty or coronary artery bypass grafting; 3) any outpatient encounter with a coronary artery disease–related diagnosis with 3 filled nitrate prescriptions; or 4) 3 outpatient encounters with coronary artery disease–related diagnoses. Diabetes status was confirmed using Kaiser Permanente Southern California’s Diabetes Case Identification Database.

The primary analyses for this study assessed the association between statins and other lipid-lowering agents and cataract. Initial analyses compared, in cases and controls, the distribution of age, race, sex, education , socioeconomic status, and statin use and duration of use. Differences in proportions were tested using the χ ² tests, supplemented by calculation of odds ratio with associated 95% confidence limits. Means were tested using the t tests. Logistic regression analysis was performed to adjust for potential confounding variables. Odds ratios were tested using the Wald test.

Results

In the calendar year 2009, 13 982 patients underwent cataract surgery in the first eye. A total of 34 049 patients had an eye examination but did not undergo cataract surgery or have the diagnosis of cataract in their record.

Table 1 shows the demographic characteristics of cases and controls. Patients who had cataract surgery were older and more likely to be white. Household income and education were similar between the 2 groups. Patients who underwent cataract surgery also appeared to have more coronary artery disease, but less diabetes.

The proportion of statin users appeared to be greater among those with cataract surgery ( Table 2 ). Longer-term users of statins also appeared to be more common among those who had cataract surgery. There was no difference in the use of other lipid-lowering drugs. Table 3 shows that the prevalence of statin use increased with age. To adjust for the relationship between statin use and age, several analyses were conducted. The first was a multivariate logistic regression analysis ( Table 4 ). After adjusting for age, female sex, race, coronary artery disease, diabetes, and smoking, statin use of 5 or more years was protective against cataract surgery, while statin use of less than 5 years raised the risk of cataract surgery. When compared to white subjects, Asian Americans were more likely, while black and Hispanic Americans were less likely, to undergo cataract surgery. Both coronary artery disease and diabetes reduced the risk of cataract surgery. The second analysis was an age-stratified multivariate analysis ( Table 5 ), in which separate logistic regression analyses were performed for those 50 to 64 years of age and for those 65 years or older. Compared to the unstratified analysis, female sex had a lower risk of cataract surgery in those 50 to 64 years of age and a higher risk in those 65 years or older. Statin use appeared to have a different effect, depending on the age group. Statin use of 5 years or more was protective against cataract surgery in those aged 50 to 64 years. Surprisingly, shorter-term use (<5 years) was associated with an increased risk of surgery in both age groups. The third analysis was a multivariate logistic regression analysis performed on cases and controls that were age-matched ( Table 6 ). In this analysis, controls were found for only 8568 cases. The other 5015 cases could not be matched because no controls could be found that were of the same age but had an eye examination and did not have a diagnosis of cataract or cataract surgery. In this analysis with a limited data set, the protective effect of longer-term statin use was still present but was not statistically significant.

TABLE 5

Multivariate Logistic Regression to Evaluate the Risk of Cataract Extraction, Stratified by Age

Factor	50−64 Years				65+ Years
		95% Wald				95% Wald
	Odds Ratio	Confidence Interval		P Value	Odds Ratio	Confidence Interval		P Value
Age	1.13	1.12	1.14	<.0001	1.12	1.11	1.13	<.0001
Female sex	0.82	0.76	0.89	<.0001	1.4	1.31	1.49	<.0001
Ethnicity/race
White	Ref				Ref
Asian	1.22	1.07	1.38	<.0001	1.44	1.27	1.62	<.0001
Black	0.56	0.49	0.65	<.0001	0.69	0.62	0.76	<.0001
Hispanic	0.71	0.64	0.79	<.001	1.08	0.99	1.18	.09
Coronary artery disease	0.82	0.76	0.89	<.0001	0.92	0.86	0.99	.02
Diabetes	1.09	1.00	1.20	.06	0.75	0.70	0.81	<.0001
Tobacco use
Never	Ref				Ref
Passive	1.01	0.60	1.71	.96	0.80	0.48	1.33	.38
Quit	1.13	1.03	1.23	.01	1.27	1.18	1.35	<.0001
Yes	1.63	1.42	1.87	<.0001	1.63	1.41	1.89	<.0001
Statin use
Never	Ref				Ref
5+ years	0.76	0.67	0.86	<.0001	0.99	0.91	1.08	.77
<5 years	1.12	1.02	1.23	.02	1.13	1.04	1.22	<.0001

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

1. Clinical Significance of Microbial Growth on the Surfaces of Silicone Tubes Removed From Dacryocystorhinostomy Patients
2. Serum Concentrations of Bevacizumab (Avastin) and Vascular Endothelial Growth Factor in Infants With Retinopathy of Prematurity
3. Reply
4. Reply

Stay updated, free articles. Join our Telegram channel

Join