Platelet Glycoprotein Receptor IIIa Polymorphism PlA2 and Ischemic Stroke Risk: The Stroke Prevention in Young Women Study

by K. Wagner, W. Giles, C. Johnson, P. Bray, P. Goldschmidt-Clermont, J. Croft, V. Brown, B. Stern, B. Feeser, D. Buchholz, C. Earley, R. Macko, R. McCarter, M. Sloan, P. Stolley, R. Wityk, M. Wozniak, T. Price, and S. Kittner

	Abstract
	Introduction
	Methods
	Results
	Discussion
	Bibliography
	Tables
	Acknowledgements

Abstract

Background and Purpose

Platelet glycoprotein IIb-IIIa, a membrane receptor for fibrinogen and von Willebrand factor, has been implicated in the pathogenesis of acute coronary syndromes, but has not been previously investigated in relationship to stroke in young adults.

Methods

We used a population-based case-control design to examine the association of the platelet glycoprotein IIIa polymorphism, PlA2, with stroke in young women. Subjects were 65 cerebral infarction cases (18 with and 47 without an identified probable etiology) 15-44 years of age from the Baltimore-Washington region and 122 controls frequency-matched by age from the same geographic area. A face-to-face interview for vascular disease risk factors and a blood sample for the PlA2 allele and serum cholesterol was obtained from each participant. Logistic regression was used to estimate the odds ratio for one or more PlA2 alleles after adjustment for other risk factors.

Results

Among cases and controls, the prevalence of one or more P1A2 alleles was 21 and 22% among blacks and 36 and 28% among whites. This genotype was significantly associated with hypertension only in black controls but otherwise not with any of the established vascular risk factors. The adjusted odds ratio for cerebral infarction of one or more PlA2 alleles was 1.1 (0.6-2.3) overall, 0.5 (0.1-7.1) among blacks, and 1.4 (0.5-3.7) among whites. For the cases with an identified probable etiology, the corresponding odds ratios were 3.0 (0.9-10.4) overall, 0.7 (0.1-7.1) among blacks and 12.8 (1.2-135.0) among whites.

Conclusions

No association was found between the PlA2 polymorphism of platelet glycoprotein IIIa in young women with stroke. However, subgroup analyses showed that the PlA2 polymorphism of platelet glycoprotein IIIa appeared to be associated with stroke risk among white women, particularly those with a clinically identified probable etiology for their stroke. Further work with an emphasis on stroke subtypes and with multiracial populations is warranted.

Introduction

Cerebral infarction in young adults is etiologically diverse¹ and a large number of risk factors have been identified, including older age, black race, hypertension, diabetes mellitus, hypercholesterolemia, and cigarette smoking. Family history of stroke has also been implicated as a risk factor,^2-4 however, few genetic risk factors for ischemic stroke have been established.

Platelet membrane glycoprotein IIb-IIIa (GPIIb-IIIa) is a platelet membrane receptor and member of the integrin family of adhesive molecules that, when activated, binds fibrinogen and von Willebrand factor, thereby promoting platelet aggregation and thrombosis.⁵ The gene encoding the GPIIIa arm of the integrin molecule is polymorphic at exon 3; the more common allele encodes a leucine (PlA1) and the less common allele encodes a proline (PlA2).⁶

Recently, the P1A2 allele of GPIIb-IIIa was reported to be an inherited risk factor for acute coronary artery events, specifically among younger white adults in some^7,8 but not all⁹ reports. There have been two reports showing no association of this polymorphism with stroke,^9,10 but neither study focused on young adults.

We postulated that the platelet polymorphism PlA2 may be a hereditary risk factor for cerebral infarction in young adults. We examined this hypothesis in a population-based case-control study in young women in the Baltimore-Washington area in analyses among both whites and blacks and for cases with and without a clinically identified probable etiology.

Methods

The Stroke Prevention in Young Women Study is a population-based case-control study in the Baltimore-Washington area initiated to study risk factors for ischemic stroke in young women. A total of 65 cases and 122 control subjects had PlA genotyping performed. Cases were women 15-44 years of age with a first cerebral infarction, identified by discharge surveillance at 59 regional hospitals and through direct referral by regional neurologists. The methods for discharge surveillance, chart abstraction, case adjudication, and assignment of probable and possible underlying causes have been described previously.^1,11 Using published criteria,¹ the case subjects were divided into two mutually exclusive groups: those with an identified probable underlying cause for their stroke (N=18) and those without an identified probable underlying cause (N=47). The group with an identified probable underlying cause for stroke was composed of atherosclerosis (greater than 60% ipsilateral stenosis)(N=6); cardiac or transcardiac emboli (N=5); nonatherosclerotic vasculopathy which included cocaine-associated cerebral infarction with no alternate cause, dissection, Takayasu's arteritis, and other vasculitides (N=4); and hematologic which included antithrombin 3 deficiency, sickle thalassemia with history of recent crisis, cancer-associated hypercoagulable state, and thrombotic thrombocytopenic purpura (N=4). One case had both an atherosclerotic and an embolic source for her stroke.

The group without an identified probable underlying cause for stroke was composed of cases with a possible underlying cause (N=25) and cases with no identified probable or possible cause (N=22). The possible causes were equivocal cardioembolic sources of embolism (N=6) which included one case with recent illicit drug use and 1 case with a possible contributing hematologic cause; lacunes (N=5) which included one case with a possible contributing role for migraine and one case with an equivocal cardioembolic source; recent illicit drug use (N=4); possible migrainous stroke (N=3) including two cases with concurrent oral contraceptive use; atherosclerosis (less than or equal to 60% stenosis)(N=3); pregnancy-associated stroke (N=2); and oral contraceptive use (N=2).

Cases were also classified as having large vessel extracranial disease or intracranial disease (N=36), small vessel disease (N=6), or indeterminate (N=23), including more than one vessel type, based on both clinical and radiologic features.

Controls were women without a history of stroke, frequency-matched by age and geographic region of residence to the cases, identified by random digit dialing.

The PlA genotyping for the gpIIIa polymorphism was performed as follows: Genomic DNA was isolated from 0.2 ml of frozen whole blood with a QIAmp blood isolation kit (Qiagen, Chatsworth, CA) .* according to the manufacturer's recommendations and eluted with 200 microliters Tris HCl pH 8.0. Five microliters of purified DNA was used in a polymerase chain reaction (PCR) containing primers (5 ' ttctgattgctggacttctctt 3 ' and 5 ' tctctccccatggcaaagagt 3 ') in a final volume of 50 microliters to yield a 266 bp DNA product.¹² One tenth (5 microliters) of the amplified DNA was digested to completion with restriction endonuclease Msp-I (Promega, Madison, WI) and electrophoresed in a 10% polyacrylamide gel to generate three genotype-related patterns as described by Weiss et al.⁷ DNA specimens corresponding to all three genotypes which had been verified by DNA sequencing were included in the lab genotyping process as controls. Genotyping was blindly performed by laboratory personnel and each sample was examined two or more times with concordant genotype results. Furthermore, independent confirmation of genotypes was obtained by blinded analysis of a subset of samples in the laboratory of P.G.-C. and P.B. using reverse dot blot hybridization and by Msp-I restriction endonuclease assay as previously described.¹³ Only four study participants were homozygous for the PlA2 allele (three cases and one control). Because of the small number of homozygotes and because there is evidence that heterozygotes are also associated with increased vascular risk,⁷ the homozygous and heterozygous groups were combined.

Potential confounders of the association between the PlA2 alleles and stroke included age, race, hypertension, diabetes mellitus, high blood cholesterol, and cigarette smoking. Hypertension and diabetes mellitus were determined by asking study participants (or their proxy if participant was unable to answer) if they had ever been told by a physician that they had the condition. Similarly, age, race, and current smoking status was determined by subject or proxy report. Cholesterol was measured according to standard practice¹⁴ and 200 mg/dl or more was considered to be a high blood cholesterol level.

T-tests were used to compare means and Chi-square tests to compare proportions. All p-values were two-sided. Adjusted odds ratios derived from logistic regression were used to determine if the presence of the PlA2 allele was associated with an increased risk for stroke after controlling for differences in age, race, hypertension, diabetes mellitus, high blood cholesterol, and cigarette smoking status.

Results

Table 1 compares women with first cerebral infarction and controls with respect to the major known vascular risk factors. Cases and controls were matched for age. Cases were more likely than controls to be of black race (49.2% vs 37.2%), were significantly more likely to have hypertension and diabetes, and tended to have higher cholesterol levels and higher cigarette smoking rates. The risk factor profile of persons with stroke due to an identified probable underlying cause was similar to that for all strokes (mean age 36.5 years, blacks 44.4%, history of high blood pressure 33.3%, diabetes mellitus 33.3%, high blood cholesterol 50%, former smoker 16.7%, current smoker 55.6%).

The cases were more likely to have traditional vascular disease risk factors than controls; if these factors were also associated with the PlA2 allele, then these factors could confound the relationship between the PlA2 allele and stroke. Table 2 examines the association between the risk factors described above and the PlA2 allele among controls, stratified by race. Hypertension was associated with the P1A2 allele among blacks only (p=.015); no other associations achieved statistical significance.

Table 3 examines the association between genotype and cerebral infarction, overall and stratified by race. Among the cases and controls, the PlA2 polymorphism was slightly more prevalent in whites (36 and 28%) than blacks (21 and 22%). After adjusting for differences in age, race, hypertension, diabetes mellitus, high cholesterol and current smoking status, the presence of at least one PlA2 allele conferred an odds ratio for stroke of 1.1 (95% CI 0.6-2.3). In addition, there was a suggestion that the PlA2 allele is a stronger risk factor for stroke among whites (OR=1.4) than among blacks (OR=0.5).

Table 4 examines the association between genotype and stroke among young women in the subset with an identified probable underlying cause of cerebral infarction (n=18). Compared to the controls, these cases had a substantially higher prevalence of the PIA2 allele (50% vs. 25%). The presence of the allele was associated with a three-fold increased risk for stroke adjusted for other factors (95% CI 0.9-10.4). This increased risk was exclusively due to the stronger association among white women (adjusted OR=12.8, 95% CI 1.2-135.0). In contrast to these findings among whites, the PlA2 allele was not associated with an increased risk of stroke among blacks (adjusted OR=0.7, 95% CI 0.1-7.1).

Cases without an identified probable underlying cause of cerebral infarction were also examined in race-stratified analyses and did not show an association between genotype and risk for stroke (data not shown).

Among cases classified as having an intracranial or extracranial large vessel stroke (N=36), the adjusted odds ratios were 0.7 (95% CI 0.2-1.8) overall, 1.3 (95% CI 0.3-4.8) for whites, and 0.1 (95% CI 0.01-1.6) for blacks. The limited number of small vessel strokes (N=6) precluded meaningful analyses in this subgroup. Among cases classified as having an indeterminate or mixed vessel type (N=23), the adjusted odds ratios were 2.4 (95% CI 0.9-6.5) overall, 3.1 (95% CI 0.7-13.3) for whites, and 1.5 (95% CI 0.3-6.9) for blacks.

Discussion

This study of stroke in young women did not show a significant association between PlA2 and all cerebral infarctions. Subgroup analyses using a previously published classification system,¹ indicated more of an effect among stroke cases with an identified probable etiology than among strokes with no identified probable etiology. PlA2 has been reported to be associated with cardiac disease^7,8, almost exclusively a large vessel process. However, our data do not support an effect predominantly among large vessel strokes. As compared to MI, stroke is a more heterogeneous process with multiple etiologies. We could identify an association between PlA2 and stroke of diverse identified causes including atherosclerosis, cardiac emboli, nonatherosclerotic vasculopathy and hematologic conditions. This suggests that the PlA2 allele could interact with other conditions predisposing to stroke, analogous to the effect of the factor V Leiden mutation on the risk of cerebral venous thrombosis.¹⁵ Women with a condition strongly predisposing to stroke may be more likely to have a stroke at an earlier age if they have the PlA2 allele as compared to women without the allele.

It is known that the PlA2 allele is less prevalent in African Americans than Caucasian populations (16% vs. 20% with one or more allele),¹⁶ but prior studies of the association of PlA2 with thrombotic events have not included blacks. Among black controls, we noted an increased prevalence of PlA2 positivity in those with hypertension. However, since our data do not support an association of the PlA2 allele with stroke in young black women, further studies will be needed to clarify the role of PlA2 in stroke and hypertension among blacks.

Prior research on the relationship of P1A2 allele to vascular disease has shown varying results. The original observation by Weiss and coworkers⁷ from Baltimore among 71 white men with myocardial infarction or unstable angina and 68 inpatient controls showed a P1A2 prevalence of 39.4% among the cases and 19.1% among the controls. The overall association was predominantly due to the effect among the 42 cases under 60 years of age, where this allele had a prevalence of 50% and was 3.6 times more frequent in patients than among controls.⁷ These observations were supported by a report by Carter from Leeds, England where the PlA2 allele was found in 50% of 24 white men with myocardial infarction before age 47 and in 27% of 45 age-matched controls.⁸ In contrast, a report based on men in the Physicians= Health Study^9,17 failed to show an association between the AlA2 allele and myocardial infarction (n=374), ischemic stroke (n=146), or venous thromboembolism (n=121). No association was evident even when the analysis was limited to cases younger than 60 years of age. Carlsson and coworkers¹⁰ found no difference in the prevalence of PlA2 or other human platelet antigen polymorphisms between 218 patients with ischemic stroke or transient ischemic attacks and 165 neurologic inpatients without acute or recent signs of cerebrovascular disease and 321 healthy blood donors. The mean age of the cases with cerebral ischemia was 62.1 years; analyses were not stratified by age.

These disparate results can be considered in the context of the several different explanations for an association of a genetic marker with disease.¹⁸ First, the marker allele may be a part of the pathologic process and define a susceptibility locus. The role of GPII-b-IIIa as a platelet membrane receptor which binds fibrinogen and von Willebrand factor provides a biological rationale for this possibility.⁵ Second, the marker allele may not cause the trait, but be in linkage disequilibrium with an unobserved "high risk" allele at a different susceptibility locus. Linkage disequilibrium is a function of the history of the population and, thus, true associations due to linkage disequilibrium can occur in one population and not in another. Third, positive associations can also occur as an artifact of population admixture. There may be confounding between unrecognized subgroups of the population, where both the marker allele frequency and the disease prevalence differ across strata of the population. Confounding may also obscure the presence of a susceptibility locus or linkage to such a locus. This potential problem may be expressed in different ways in different populations. Therefore, the disparate results among studies of the association of the PlA2 allele with disease may be due to linkage disequilibrium, confounding by population admixture, or to age differences. Our population-based case control design with analysis stratified by race was intended to minimize the problem of population admixture.

A limitation of our study is the small sample size which increases the likelihood of both Type 1 and Type 2 errors. As stroke in the young is uncommon, it is difficult to obtain a large population based patient group for a case control study. Ours is the only study of the PlA2 allele in a multiracial population of young women. Our results are promising as the wide confidence intervals do not exclude the possibility of a odds ratio in the range of 2-4 for the overall group. Similarly, the strong effect among the subgroup of white women with an identified probable cause, while statistically significant, also had wide confidence intervals and will require replication.

The formation of a platelet clot requires the binding of fibrinogen and von Willebrand factor to its receptor, glycoprotein IIb-IIIa on the platelet surface.⁵ Antiplatelet therapy has been a mainstay of both primary and secondary stroke prevention. This therapy has included aspirin, which inhibits platelet cyclooxygenase and thromboxane A2 production, and ticlopidine, which inhibits ADP activation of glycoprotein IIb/IIIa.^19,20 Other antiplatelet agents which selectively inhibit the glycoprotein receptor have been developed.²¹ PlA2 is a highly prevalent polymorphism in both whites and African Americans. Confirmation that this allele affects susceptibility to stroke would raise the prospect of stroke prevention efforts specific to genotype status. Further work in this area with an emphasis on stroke subtypes and with multiracial populations is warranted.

Bibliography

1. Johnson CJ, Kittner SJ, McCarter RJ, et al. Interrater reliability of an etiologic classification of ischemic stroke. Stroke 1995;26:46-51.
2. Khaw K, Barrett-Connor E. Family history of stroke as an independent predictor of ischemic heart disease in men and stroke in women. American Journal of Epidemiology 1986;123:59-66.
3. Welin L, Scardsudd K, Wilhelmsen L, Larsson B, Tibblin G. Analysis of risk factors for stroke in a cohort of men born in 1913. New Engl J Med 1987;521-526.
4. Carrieri PB, Orefice G, Maiorino A, Provitera V, Balzano G, Lucariello A. Age-related risk factors for ischemic stroke in Italian men. Neurepidemiology 1994;13:28-33.
5. Phillips DR, Charo IF, Parise LV, Fitzgerald LA. The platelet membrane glycoprotein IIb-IIIa complex. Blood 1988;71:831-843.
6. Newman PJ, Derbes RS, Aster RH. The human platelet alloantigens PIA1 and PIA2 are associated with a leucine 33/ proline 33 amino acid polymorphism in membrane glycoprotein IIIa and are distinguishable by DNA typing. J Clin Invest 1989;83:1778-1781.
7. Weiss EJ, Bray PF, Tayback M, Schulman SP, Kickler TS, Becker LC, et al. A polymorphism of a platelet glycoprotein receptor as an inherited risk factor for coronary thrombosis. The New England Journal of Medicine 1996;334:1090-1094.
8. Carter AM, Ossei-Gerning N, Grant PJ. Platelet Glycoprotein IIIa P1A polymorphism in young men with myocardial infarction. Lancet 1996;348:485-486.
9. Ridker PM, Hennekens CH, Schmitz C, Stampfer MJ, Lindpainter K. P1A1/A2 polymorphism of platelet glycoprotein IIIa and risks of myocardial infarction, stroke, and venous thrombosis. Lancet 1997;349:385-388.
10. Carllsson LE, Greinacher A, Spitzer C, Walther R, Kessler C. Polymorphisms of the human platelet antigens HPA-1, HPA-2, HPA-3, andHPA-5 on the platelet receptors for fibrinogen (GPIIb/IIIa), von Willebrand factor (GPIb/IX), and collagen (GPIa/Iia) are not correlated with an increased risk of stroke. Stroke 1997;28:1392-1395.
11. Kittner SJ, Stern BJ, Feeser BR, Hebel JR, Nagey DA, Buchholz DW, et al. Pregnancy and the risk of stroke. The New England Journal of Medicine 1996;335:768-774.
12. Jin Y, Dietz HC, Nurden A, Bray PF. Single-strand conformation polymorphism analysis is a rapid and effective method for identification of mutations and polymorphisms in the gene for glycoprotein IIIa. Blood 1993;82:2281-2288.
13. Bray PF, Jin Y, Kichler T. Rapid genotyping of the five major platelet alloantigens by reverse dot-blot hybridization. Blood 1994;84:4361-4367.
14. Anonymous Fundamentals of Clinical Chemistry. Philadelphia, PA: W.B. Saunders, Co. 1976.
15. Zuber M, Toulon P, Marnet L, Mas JL. Factor V Leiden mutation in cerebral venous thrombosis. Stroke 1996;27(10):1721-1723.
16. Kim H, Jin Y, Kickler TS, Blakemore K, Kwon OH, Bray PF. Gene frequencies of the five majors human platelet antigens in African-American, white and Korean populations. Transfusion 1995;35:865-867.
17. Ridker PM. P1A1/A2 polymorphism of platelet glycoprotein IIIa and risk of cardiovascular disease. Lancet 1997;349:1100
18. Lander ES, Shork NJ. Genetic dissection of complex traits. Science 1994;265:2037-2048.
19. Fitzgerald GA, Meagher EA. Antiplatelet drugs. Eur J Clin Invest 1994;24 Suppl:46-49.
20. Haynes RB, Sandler RS, Larson EB. A critical appraisal of ticlopidine, a new antiplatelet agent. Effectiveness and clinical indications for prophylaxis of atherosclerotic events. Arch Intern Med 1992;152:1376-1380.
21. Lefkovits J, Plow EF, Topol EJ. Platelet glycoprotein Ib/IIIa receptors in cardiovascular medicine. New Engl J Med 1995;332:1553-1559.

Table 1. Baseline characteristics of study participants. The Stroke Prevention in Young Women Study.

* T-tests are used to compare means and chi-square test to compare proportions. P-values are two-sided.

Table 2. Prevalence of one or more PlA2 alleles according to selected characteristics. The Stroke Prevention in Young Women Study.

Chi-square tests are used to compare proportions. P-values are two-sided.

Table 3. Presence of one or more A2 alleles and risk for cerebral infarction, Stroke Prevention in Young Women Study.

Adjusted model controls for age, race, hypertension, diabetes, high blood cholesterol, and cigarette smoking. Race-specific models do not adjust for race.

Table 4. Presence of one or more A2 alleles and risk for cerebral infarction, in cases with an identified probable cause for stroke. The Stroke Prevention in Young Women Study.

Adjusted model controls for age, race, hypertension, diabetes, high blood cholesterol, and cigarette smoking. Race-specific models do not adjust for race.

Acknowledgements

We are indebted to Chad H.Richardson for his outstanding technical assistance and to the following members of the Stroke Prevention in Young Women research team for their dedication: Anne Epstein, James Gardner, Mary Keiser, Ann Maher, Jennifer Rohr, Mary J. Seipp, Susan Snyder, Mary J. Sparks, and Nancy Zappala.

The authors would like to acknowledge the assistance of the following individuals who have sponsored the Stroke Prevention in Young Women Study at their institution: Frank Anderson, M.D.; Clifford Andrew, M.D., Ph.D.; Christopher Bever, M.D.; Nicholas Buendia, M.D.; Young Ja Cho, M.D.; James Christensen, M.D.; Remzi Demir, M.D.; Terry Detrich, M.D.; John Eckholdt, M.D.; Nirmala Fernback, M.D.; Jerold Fleishman, M.D.; Benjamin Frishberg, M.D.; Stuart Goodman, M.D., Ph.D.; Norman Hershkowitz, M.D. Ph.D.; Luke Kao, M.D. Ph.D.; Mehrullah Khan, M.D.; Ramesh Khurana, M.D.; John Kurtzke, M.D.; William Leahy, M.D.; William Lightfoote II M.D.; Bruce Lobar, M.D.; Michael Miller, M.D., Ph.D.; Harshad Mody, MBBS; Marvin Mordes, M.D.; Seth Morgan, M.D.; Howard Moses, M.D.; Sivarama Nandipati, M.D.; Mark Ozer, M.D.; Roger Packer, M.D.; Thaddeus Pula, M.D.; Philip Pulaski, M.D.; Naghbushan Rao, M.D.; Marc Raphaelson, M.D.; Solomon Robbins, M.D.; David Satinsky, M.D.; Elijah Saunders, M.D.; Michael Sellman, M.D.,Ph.D.; Arthur Siebens, M.D.(deceased); Harold Stevens, M.D., Ph.D.; Dean Tippett, M.D.; Roger Weir, M.D.; Michael Weinrich, M.D.; Richard Weisman, M.D.; Don Wood, M.D.(deceased); and Mohammed Yaseen, M.D.

In addition, the study could not have been completed without the support from the administration and medical records staff at the following institutions: Maryland - Anne Arundel Medical Center, Atlantic General Hospital, Bon Secours Hospital, Calvert Memorial Hospital, Carroll County General, Church Hospital Corporation, Doctors Community Hospital, Fallston General Hospital, Franklin Square Hospital Center, Frederick Memorial Hospital, The Good Samaritan Hospital of Maryland, Inc., Greater Baltimore Medical Center, Harbor Hospital Center, Harford Memorial Hospital, Holy Cross Hospital, Johns Hopkins Bayview Inc., The Johns Hopkins Hospital, Howard County General Hospital, Inc., Kennedy Krieger Institute, Kent and Queen Anne Hospital, Laurel Regional Hospital, Liberty Medical Center, Inc., Maryland General Hospital, McCready Memorial Hospital, Memorial Hospital at Easton, Mercy Medical Center, Montebello Rehabilitation Hospital, Montgomery General Hospital, North Arundel Hospital, Northwest Hospital Center, Peninsula Regional Medical Center, Physician's Memorial Hospital, Prince George's Hospital Center, Saint Agnes Hospital, Saint Joseph Hospital, Saint Mary's Hospital, Shady Grove Adventist Hospital, Sinai Hospital of Baltimore, Southern Maryland Hospital Center, Suburban Hospital, The Union Memorial Hospital, University of Maryland Medical System, Department of Veterans Affairs Medical Center in Baltimore, Washington Adventist Hospital, and Washington County Hospital; Washington D.C. - Children's National Medical Center, District of Columbia General Hospital, The George Washington University Medical Center, Georgetown University Hospital, Greater Southeast Community Hospital, Hadley Memorial Hospital, Howard University Hospital, National Rehabilitation Hospital, Providence Hospital, Sibley Memorial Hospital, Veterans Affairs Medical Center, and The Washington Hospital Center; Pennsylvania - Gettysburg Hospital, and Hanover General Hospital.