LETTER TO THE EDITOR Central retinal artery occlusion with atrial fibrillation or atrial flutter Aditya Uppuluri 1 & Sohil Bhagat 2 & Marco A. Zarbin 1 & Neelakshi Bhagat 1 Received: 28 October 2020 /Revised: 28 October 2020 /Accepted: 6 November 2020 # Springer-Verlag GmbH Germany, part of Springer Nature 2020 Dear Editor, In this present study, the 2002–2014 National Inpatient Sample (NIS) Database was used to perform a retrospective cross-sectional case-control study to identify comorbidities associated with a diagnosis of central retinal artery obstruction (CRAO) in patients with atrial fibrillation and atrial flutter (AFF) over 65 years of age [1, 2] While previous studies have established that AFF increases the risk of CRAO [3, 4], this is the first study to utilize a national dataset to identify additional risk factors in this at-risk population. The International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) was used to identify cases and controls, systemic conditions, and ocular findings. The case group (739 patients) comprised patients over 65 years of age who were admitted to the hospital for CRAO (ICD-9 code 362.31) and a concurrent diagnosis of AFF. The age- and sex-matched control group (3689 patients) included randomly selected patients over the age of 65 years with AFF and without retinal artery occlusion. Statistical anal- ysis was performed using IBM SPSS 23 and R package ver- sion 3.4.3. CHADS-VASC scores were calculated as per the guidelines put forth by Lip and colleagues [5]. In the 2002–2014 NIS Database, there were 739 (weighted) cases and 3689 (weighted) controls. The cases and controls were age- and gender-matched. The average ages of cases and controls were 79.2 years and 79.4 years (p = 0.452), respectively. Men comprised 55.0% of con- trols and 55.6% of cases (p = 0.759). Table 1 presents the results of our chi-square analysis and independent samples t testing. There was no significant dif- ference in the CHADS-VASC scores of the controls and cases (3.90 vs. 4.00; p = 0.055). Carotid stenosis, hyperlipidemia, hypertension, history of stroke (hemorrhagic and ischemic), and glaucoma were more prevalent in patients with CRAO than in those without CRAO (p < 0.05). Cases had a shorter duration of hospitalization com- pared to controls (3.60 days vs. 5.90 days; p < 0.001). Cases had a higher average cost per day ($9407) than controls ($7989) (p < 0001). With regard to acute com- plications, 5.7% of patients with CRAO developed non- hemorrhagic vaso-occlusive cerebral stroke during hos- pitalization, compared to 1.0% of controls (p < 0.001). Controls had a higher rate of in-hospital myocardial infarctions, 2.6%, than cases, 1.2% (p = 0.024). In the control group, 171 (4.6%) patients died during hospital- ization; no patients admitted for CRAO died during hospitalization. Carotid stenosis (OR = 23.77; CI = 16.16–35.00), hyper- lipidemia (OR = 1.72; CI = 1.40–2.11), hypertension (OR = 1.81; CI = 1.42–2.31), and history of hemorrhag- ic or ischemic stroke (OR = 2.57; CI = 1.82–3.63) were associated with an increased risk of developing CRAO in patients with AFF (Table 2). Limitations to this study include its retrospective nature, its reliance of ICD-9 codes for diagnoses, the lack of longitudinal follow-up, and overrepresentation of chronic medical condi- tions associated with using an inpatient dataset; however, these limitations are not unique to the NIS Database or to this study. In our case-control study of the NIS Database, we found that carotid stenosis, hyperlipidemia, hypertension, and history of stroke (ischemic and hemorrhagic) were all associated with further increased risk of CRAO in the at-risk population with AFF. Ischemic strokes were five times more common in AFF patients admitted for CRAO and occurred in 5.7% of AFF patients within 6 days of hospital admission. More recent practice patterns recommend a stroke work-up on newly diagnosed CRAOs[6, 7]. Further pro- spective, well-designed studies can address incidence, appropriate management, and optimal anticoagulation to prevent ischemic strokes in AFF patients with new CRAO. * Neelakshi Bhagat bhagatne@njms.rutgers.edu 1 Department of Ophthalmology and Visual Sciences, Rutgers New Jersey Medical School, 90 Bergen Street Doctor Office Center, Suite 6100, Newark, NJ 07103, USA 2 A.T. Still University, 800 West Jefferson Street, Kirksville, MO 63501, USA Graefe's Archive for Clinical and Experimental Ophthalmology https://doi.org/10.1007/s00417-020-05014-x