History of Hollenhorst Plaques
Hollenhorst plaques (Figure 1) refer to yellow, orange refractile cholesterol emboli which occur at retinal arteriole bifurcations arising from carotid or aortic arch atheromatous disease. Independently, Hollenhorst1 and German investigators,2 Witmer and Schmid, described these retinal plaques in 1958. Hollenhorst’s later work established the clinical significance of the plaques as markers of cerebrovascular disease3 and increased mortality.4 His observations and experimental work demonstrated that the plaques contained cholesterol.5
Biography of Hollenhorst
Robert W. Hollenhorst (Figure 2) was born in St. Cloud, MN, on August 12, 1913. Although singing was his first career interest (“it paid a good bit of my way through school”), he eventually attended the University of Minnesota Medical School. He subsequently served in the US Army Medical Corps during World War II and was awarded a Bronze Star for keeping his division’s scrub typhus death rate significantly lower than that of other South Pacific regions. After leaving the military, he began his residency in ophthalmology at Mayo Clinic in Rochester, MN, followed by a staff position. Before his landmark publications on retinal cholesterol emboli, Hollenhorst described a series of patients who, after anesthesia, experienced unilateral blindness related to increased ocular pressure and lowered blood pressure,6 which became known as Hollenhorst syndrome.7 He also authored important papers on giant cell arteritis8 and the use of cortisone for eye disease.9 In addition to Hollenhorst’s detailed examinations, the unique practice environment at Mayo Clinic during his tenure in neuroophthalmology fostered multidisciplinary observations as all patients seen in neurology also saw neuroophthalmology. This allowed Hollenhorst to see a large number of patients with carotid occlusive disease. He worked at Mayo Clinic for 33 years. In 1986, the American Ophthalmological Society awarded Hollenhorst the prestigious Howe Medal. He died on January 10, 2008, in Rochester, MN.
Original Descriptions of Hollenhorst Plaques
In 1927, Butler provided an early description of retinal emboli, describing a bright embolus in the inferior temporal arteriole.10 Several early observations of retinal embolism include those of Gowers in 1875, Minton in 1936, Duke-Elder in 1940, and Fisher in 1952 and 1958, but few descriptions were similar to the bright plaques described by Hollenhorst.3 In 1958, Hollenhorst reported retinal arteriole plaques in 4/86 patients with carotid artery disease. He noted that the emboli tended to occur ipsilateral to the side with carotid disease.1 Also in 1958, Witmer and Schmid reported the case of a 54-year-old male who developed a shiny, yellow object in a left retinal artery. They speculated that the lesion was a cholesterol embolus, which had been dislodged from the carotid artery after the patient struggled with buttoning his collar.2 In 1960, Hollenhorst expanded on his previous findings, describing the plaques as bright and noting that they were mobile with manual pressure. He also mentioned the appearance of plaques with carotid artery surgery.11 In 1961, Hollenhorst published his landmark article, “Significance of Bright Plaques in Retinal Arterioles” in JAMA where he observed bright, orange-yellow plaques in 11% (27/235) of patients with carotid disease and 4% (4/93) with vertebrobasilar disease.3 Through careful observation, Hollenhorst identified the carotid artery as an important source of bright retinal plaques, proposed cholesterol as the composition, and identified several cases that occurred during or shortly after carotid surgery.3 He also pointed out that the plaques rarely blocked the arterioles and noted they could migrate to distal bifurcations, subsequently breaking into smaller fragments.3
Hollenhorst’s Experimental Investigations and Follow-Up Studies
To provide evidence for the cholesterol basis for bright plaques, Hollenhorst, along with cerebrovascular neurologist Dr Jack Whisnant, injected cholesterol crystals and suspended crystals from human atheromatous material into the carotid arteries of dogs and Macaca rhesus monkeys. Both cholesterol crystals and atheromatous material reproduced retinal bright plaques similar in appearance at arteriole bifurcations as seen in humans. This strongly supported Hollenhorst’s suggestion of cholesterol as the origin.5 This study served as Hollenhorst’s thesis to become a member of the American Ophthalmological Society. Interestingly, Whisnant later described the phenomenon of retinal claudication (unilateral loss of vision in bright light; “Whisnant’s phenomenon”) as a manifestation of carotid occlusive disease. Later in 1963, other investigators pathologically proved the cholesterol ester composition of bright plaques in an autopsy of a patient who had bright plaques in retinal arterioles and died after carotid endarterectomy.12 Based on these findings, the authors distinguished the bright plaques composed of cholesterol described by Hollenhorst with the white plugs composed of platelets, thrombi, and hyalin described by Fisher, Russell, and McBrien.12
In 1966,4 Hollenhorst published observations on 208 consecutive additional patients with retinal cholesterol plaques over a 5-year period, noting an association with diabetes mellitus, hypertension, and smoking. He noted the high number with atheromatous disease (63% with TIA or stroke, 59% with coronary artery disease, 42% with peripheral vascular disease). Although no formal follow-up was ascertained, Hollenhorst observed that 23 of the patients died, most often from myocardial infarction. Key observations related to cerebrovascular disease included a temporal relationship between the appearance of retinal cholesterol crystals and the episodes of cerebral ischemia and the appearance of cholesterol emboli in the eye ipsilateral to the symptomatic carotid. In a follow-up study of 205/208 patients with retinal cholesterol embolism, Hollenhorst demonstrated that survival in patients with retinal embolism was worse than expected: 13% less at 1 year and 40% less at 8 years.13 Hollenhorst’s detailed observations about cholesterol emboli as biomarkers for systemic cerebrovascular and cardiovascular disease with increased mortality set the stage for more recent publications.
More Recent Studies of Retinal Emboli
The advent of noninvasive carotid artery imaging has allowed confirmation that Hollenhorst plaques are predictive of carotid artery occlusive disease, with ≈25% having carotid stenosis of >40%.14
A 1995 VA medical center study demonstrated that Hollenhorst plaques were a significant risk factor for ischemic stroke.15 In the population-based Blue Mountains Eye Study, retinal emboli were found in 1.4% of participants over age 49 years, with an increasing prevalence by decade. Eighty percent were cholesterol emboli. Similar to Hollenhorst’s observation in 1966, the presence of retinal emboli was associated with hypertension, systemic arterial disease, and smoking.16 In the Beaver Dam Eye Study, a population-based study in Wisconsin, the 10-year incidence of retinal embolism was 10%. Retinal emboli were also associated with an increased risk of death caused by stroke, although subtypes of emboli were not specified.17 In a pooled analysis of mortality over 10 to 12 years from the Beaver Dam Eye Study and Blue Mountains Eye Study, the cumulative mortality was 56% with retinal emboli and 30% without retinal emboli, with stroke-related mortality specifically increased.18
Retinal cholesterol emboli serve as biomarkers of carotid artery disease and predict co-occurring cardiovascular disease and increased mortality. Hollenhorst provided early descriptions of these bright plaques. The unique practice at Mayo Clinic where all neurology patients also saw neuroophthalmology provided the basis for identifying the association with carotid artery disease. Hollenhorst performed key experiments proving his prior speculation that the composition of the bright plaques was cholesterol. Through detailed baseline examinations and follow-up, Hollenhorst demonstrated that the plaques had significant predictive value. The eponym, Hollenhorst plaques, recognizes this contribution.
- Received November 17, 2014.
- Revision received December 16, 2014.
- Accepted December 22, 2014.
- © 2015 American Heart Association, Inc.
- Butler TH
- Mitchell P,
- Wang JJ,
- Li W,
- Leeder SR,
- Smith W
- Wang JJ,
- Cugati S,
- Knudtson MD,
- Rochtchina E,
- Klein R,
- Klein BE,
- et al