Progression of Diabetic Retinopathy in the Hypertension Intervention Nurse Telemedicine Study

Interventions to improve glycemic control reduce the progression of diabetic retinopathy.¹ Additionally, pharmacologic reduction of blood pressure (BP) in patients with diabetes mellitus and poorly controlled hypertension reduces the risk of worsening diabetic retinopathy.² Despite pharmacologic improvements, hypertension remains poorly controlled in approximately half of Americans with it.³ The Hypertension Intervention Nurse Telemedicine Study (HINTS) investigated a telemedicine-mediated medication management and behavioral intervention for hypertension. Participants in HINTS (including nondiabetic individuals) with poor BP control who received combined medication and behavioral management demonstrated mean decreases in systolic BP of 15 mm Hg and 8 mm Hg at 12 and 18 months, respectively.⁴ The purpose of the current analysis was to determine whether the interventions influenced the progression of diabetic retinopathy in participants with diabetes.

Eligibility criteria for HINTS included veterans having hypertension with inadequate BP control (average BP in past 12 months >140/80 mm Hg). Patients were excluded for hemodialysis or a creatinine level greater than 2.5 mg/dL (to convert to micromoles per liter, multiply by 88.4). All participants received home BP monitors. Participants were randomized to the following: (1) usual care; (2) nurse-administered behavioral intervention; (3) nurse-administered medication management; or (4) a combination of the 2 interventions (Table). In institutional review board–approved secondary analyses of diabetic participants, a single chart abstractor (K.W.M.) determined the presence and severity of diabetic retinopathy at baseline and the most recent follow-up, as recorded by the eye care provider of record in the electronic chart. Eligible participants had at least 1 documented dilated examination prior to or within 2 months of enrollment and at least 1 subsequent dilated examination 365 days or later following enrollment. Progression was defined as one or both eyes moving 1 or more steps along the spectrum of retinopathy: none, mild nonproliferative, moderate nonproliferative, severe nonproliferative, proliferative, and visual acuity less than 20/60 in the better-seeing eye. Logistic regression was used to examine the association between progression of diabetic retinopathy and the intervention group, adjusting for baseline hypertension control and duration of follow-up.

Of the 593 veterans enrolled in HINTS, 252 were identified as diabetic. Of the 194 participants meeting the additional criteria for documented eye examinations, 58 (30%) had diabetic retinopathy at baseline and 65 (34%) experienced progression of retinopathy in at least 1 eye at follow-up (mean [SD] follow-up, 1255 [344] days; median follow-up, 1310 days). After controlling for duration of follow-up, the odds of diabetic retinopathy progression were significantly greater among participants receiving usual care than among participants receiving medication management, either alone or in combination with behavioral management (odds ratio = 2.16; 95% CI, 1.03-4.52; P = .04), but not different from the group receiving behavioral management alone (odds ratio = 0.88; 95% CI, 0.40-1.95; P = .84).

A nurse-administered medication management program facilitated by home BP monitoring was associated with decreased risk of progression of retinopathy in diabetic individuals with comorbid hypertension. Although intensive pharmacologic control of systolic BP (<120 mm Hg) may not be protective against worsening diabetic retinopathy,⁵ the more moderate reduction in systolic BP achieved in HINTS via the nurse-administered telemedicine program⁴ (8 mm Hg lower in the combined intervention group vs the control group) may offer greater benefit. Telemedicine also has the potential advantage of reaching patients with suboptimal access to traditional care. Diabetic individuals with renal disease were excluded from HINTS, likely limiting the population to diabetic individuals with less severe retinopathy. As such, the results of this study are consistent with other work suggesting that BP control is most helpful in controlling diabetic retinopathy in early disease.⁶ Although as a secondary analysis this study is limited in the ability to support causation (for example, less progression of retinopathy in the combined intervention group may be related to the intervention but not a direct result of decreasing BP), the potential benefit of telemedicine interventions targeting BP control in patients at risk for diabetic eye disease is encouraging and warrants further study.

Corresponding Author: Dr Muir, Department of Ophthalmology, Duke University Medical Center, PO Box 3802, Durham, NC 27710 (kelly.muir@duke.edu).

Published Online: May 23, 2013. doi: 10.1001/jamaophthalmol.2013.81

Conflict of Interest Disclosures: None reported.

Trial Registration: clinicaltrials.gov Identifier: NCT00129103.