Objectives: Currently, headache, nausea/vomiting, visual changes, and altered mental status are accepted as indications for the evaluation of hydrocephalus in children; while dementia, gait apraxia, and urinary incontinence remain indications in the elderly. The clinical presentation of hydrocephalus in young and middle-aged adults remains poorly described. Hence, middle-aged patients with mild gait, cognitive, or urinary symptoms unaccompanied by clear exam findings often remain undiagnosed and untreated.
Methods: We report the clinical presentation, treatment, and outcomes of 46 adults (ages 16-55 years) presenting with congenital, acquired, or idiopathic hydrocephalus with imaging-documented ventriculomegaly and elevated CSF pressure.
Results: Primary symptoms were related to gait (70%), cognition (70%), urinary urgency (48%), and headaches (56%). Eighty-four percent complained of impaired job performance. The exam findings were subtle or absent (no gait apraxia, minor gait changes in 42.9%, mildly abnormal Mini Mental State exams in only 14.3%, and incontinence in only 3.6%). Twenty-nine patients underwent ventriculoperitoneal (VP) shunting, and 11 endoscopic third ventriculostomy, of whom six subsequently required a VP shunt. Symptomatic improvement was observed in 93% of patients 16±11 months after shunting (56% complete resolution, 37% partial resolution). Patients had been followed for their symptoms an average of 6 years (range, 1-30) prior to diagnosis.
Discussion: We propose that there exists a clinically distinct syndrome of hydrocephalus in young and middle-aged adults (SHYMA) that comprises hydrocephalus of all etiologies. SHYMA is characterized by complaints of impaired gait, cognition, bladder control, and headaches, with a discrepancy between the prominence of symptoms and the subtlety of clinical signs. Despite the subtlety of clinical signs, CSF diversion treatment is effective at resolving symptomatology. [Neurol Res 2005; 27: 540-547]
Keywords: Hydrocephalus; adults ages 18-55 years; headache; syncope; ventriculoperitoneal shunt; endoscopic third ventriculostomy
Hydrocephalus is usually considered to present at the extremes of age-among infants and the elderly-and it has a characteristic syndrome for each of these age groups. In infants, hydrocephalus presents with head enlargement and symptoms ranging from irritability to seizures, and the prognosis ranges from normal development to severe developmental delay1,2. In children, hydrocephalus often presents with headache, nausea/ vomiting, visual changes, or altered mental status. Early diagnosis and shunting results in resolution of symptoms. Symptomatic hydrocephalus of the elderly presents with dementia, gait apraxia and imbalance, and urinary incontinence, widely known as normal pressure hydrocephalus (NPH), and the prognosis ranges from complete restoration of function to gradual deterioration3.
Less commonly known, and less commonly recognized, is hydrocephalus presenting in young and middle-aged adults4,5. The syndrome of late-onset idiopathic aqueductal stenosis (LIAS) has been described as having one presentation for young adults (headaches) and another for older adults (NPH symptoms)6. The syndrome of long-standing overt ventriculomegaly (LOVA) was originally characterized by congenital aqueductal stenosis with headaches, 'subnormal IQ', and macrocephaly, with facultative symptoms of urinary incontinence or gait disturbance7, and was subsequently revised to include patients 'with or without subnormal IQ'7.
The presentation of hydrocephalus in young and middle-aged adults is often atypical or subtle enough to not be recognized as clinical evidence of hydrocephalus. Furthermore, the clinical presentation of hydrocephalus strictly in young and middle-aged adults has yet to be characterized. Hence, middle-aged patients with mild gait, cognitive, or urinary symptoms unaccompanied by clear exam findings often remain undiagnosed and untreated. The purpose of this manuscript is to describe the syndrome of hydrocephalus in young and middle-aged adults (SHYMA).
We report 46 patients referred to the Johns Hopkins Hospital between 1993 and 2003. Inclusion criteria included decompensated congenital hydrocephalus (DCH), acquired hydrocephalus (AH), or idiopathic hydrocephalus (IH) in the age range of 16-55 years. All patients completed a standard questionnaire regarding their symptoms, medical history, and social history, and all were evaluated by a single neurologist (M.A.W.). The diagnosis of symptomatic hydrocephalus was made on the basis of history, signs and symptoms, computed tomography (CT) or magnetic resonance images (MRI) documented hydrocephalus, elevated intracranial pressure (ICP), and response to external cerebrospinal fluid drainage.
Based on medical history and physical exam, patients were classified as having DCH, AH, or IH. Patients were included in the DCH group if their head circumference was at the 97th percentile or more for sex and height8. Those in the AH group had medical histories including head trauma, meningitis, encephalitis, or brain tumor. Head trauma was considered significant if it resulted in loss of consciousness, hospitalization, or was repetitive (e.g. boxing). Patients not meeting the criteria for the two previous groups were considered to have IH.
The age of symptom onset was determined by asking patients, family, or friends when they first noticed symptoms, or if possible, by obtaining medical records. The age of diagnosis was defined either as the point at which the patient's ventriculomegaly was associated with their most recent symptoms, or when physiological testing such as ICP monitoring or CSF drainage supported the diagnosis of hydrocephalus9. The time-to-diagnosis (TTD) was determined by calculating the difference between the age of diagnosis and the age of symptom onset. Based on assessment of test results and severity of symptoms, including impact on job performance, patients were advised about treatment options. Patients who underwent shunt surgery or endoscopic third ventriculostomy (EW) were evaluated by a neurosurgeon (D.R.) at 3 and 6 months, and yearly thereafter.
The outcomes were assessed by determining the extent of symptom resolution, and if applicable, ability to return to employment. Complete improvement was defined as complete resolution of all presenting symptoms plus the ability to return to employment (if applicable). Partial improvement was defined as complete resolution of at least one of the presenting symptoms. Patients were classified as having no improvement if all of their pretreatment symptoms persisted at follow-up.
Data were analyzed using SPSS(TM). The analysis of continuous variables included mean, median, range, and ANOVA. Categorical variables were analyzed using Pearson's chi-squared and Fisher's exact test with the level of significance set at p≤0.05.
We evaluated 46 patients (23 men; 23 women) between 1993 and 2003. Seven patients (15%) had DCH, 19 (41%) had AH, and 20 (43%) had IH. The cause of AH included head trauma (n=8), meningitis/encephalitis (n=4), surgically resected brain tumor (n=2), and aqueductal stenosis from cyst (n=2), cavernoma (n=2), or unruptured aneurysm (n=1). Hydrocephalus was communicating in 16 (35%) cases and obstructive in 30 (65%). The head circumference ranged from the 25th to the 90th percentiles in the AH group, from the 50th to the 80th percentiles in the IH group, and was more than 97th percentile in the DCH group. The highest level of education obtained was a graduate degree for 11 (24%), undergraduate degree for 14 (30%), high school for 17 (37%), and less than high school for four (9%) patients. Other medical diagnoses were hypertension (21.4%), psychiatric illness (21.4%), diabetes mellitus (10.7%), and epilepsy (8.0%). Table 1 describes our patient population.
The mean age at symptom onset was 35 years (median 36; range 16-53). The mean TTD for all patients was 6 years (range 0.3-34.6 years). At the time of symptom onset, 35 patients were employed. Symptoms at the time of diagnosis were related to gait in 32 (70%), cognition in 32 (70%), bladder control in 22 (48%). Symptoms classically associated with hydrocephalus were observed less frequently, i.e. visual changes in nine (19%), nausea/vomiting in six (13%), and alteration in consciousness in one (2%). Other symptoms included chronic headaches that were unrelieved by analgesics in 26 (56%), change in sleep patterns in 12 (26.0%), and recurrent syncope in seven (15%) patients.
Disturbances in gait were most frequently described as clumsiness, difficulty on uneven surfaces, and difficulty with stairs. Physical findings of subtle gait abnormalities (widened base or stance, shortened stride length, impaired tandem stance or walk, or increased number of steps for a 180° turn) were seen in 21/46 (45%) patients. Overt gait apraxia was absent, aside from one patient who was so severely disabled after 20 years of symptoms as to require a wheelchair, and another who had developmental delay and was unable to bear weight due to treatment of spastic diplegia with intrathecal baclofen.
Cognitive symptoms focused primarily on mental 'dullness', poor organizational skills, or dependence on lists, rather than frank dementia. Of the 35 employed patients surveyed, 22 of the 25 (84%) complained that their symptoms resulted in a significant decrease in job performance. Only six of the 46 (13%) patients had abnormal exams10 at the initial evaluation. One patient was obtunded; one had severe mental retardation; one had a long-time learning ability, and the other three had 10-20 years of cognitive decline.
The chief urinary symptoms were frequency and urgency. Only three (6%) patients had urinary incontinence.
Forty (87%) patients were treated: 29 with ventriculoperitoneal (VP) shunts, 11 with ETV. Six patients (one DCH, three IH, two AH) did not choose to undergo surgical intervention. Of the 29 patients treated with VP shunts, nine underwent a single subsequent revision, six underwent two revisions; and four patients underwent more than two revisions during the follow-up period. Of the 11 patients treated with ETV alone, six required subsequent shunt insertion.
The mean duration of follow-up after the initial surgery was 1 6 months (range 4-46). One patient was lost to follow-up after surgery. Symptomatic improvement was seen in 36 (93%) patients after CSF shunting. Complete improvement was seen in 22 out of 39 patients (56%), and partial improvement in 14 out of 39 (38%). Three (7%) patients had no change even after shunt revision. No patients had worsening of symptoms during the initial follow-up. No statistical difference in outcome was found with regard to sex, age of symptom onset, time-to-diagnosis, diagnosis type, or type of symptoms.
Case 1 (IH, communicating)
A 41-year-old man had dizziness, dysequilibrium, urinary frequency, and worsening memory difficulty and poor organizational skills (including inability to follow a musical score) for ~6 months that interfered with his job as a music teacher to the extent that he received a poor job performance review. He had 7 years of bifrontal headaches that had worsened in the 6 months prior to evaluation, and occasionally disturbed his sleep. Lumbar punctures on two occasions showed pressures of 530 mm H2O and 1 70 mm H2O. A trial of acetazolamide by his referring physician failed to relieve his symptoms. MRI revealed ventricular dilation with an enlarged third ventricle. Physical exam was normal. ICP monitoring showed B-waves (Figure 7), and his symptoms improved with controlled CSF drainage. A PS Medical Delta valve (level 1.0) was inserted and he had a good recovery. He returned to work with excellent job performance reviews, has no headaches, no urinary symptoms, and walks 1 mile/day.
Case 2 (DCH, obstructive)
A 37-year-old woman had progressive gait difficulty (most notably had difficulty climbing stairs), memory loss, and urinary frequency and urgency for 2 years. Her head circumference was above the 97th percentile. Hydrocephalus had been diagnosed in infancy and treated with a VP shunt, but she received no follow-up for hydrocephalus. ICP monitoring revealed unequivocal elevation of the ICP. After the insertion of a PS Medical Delta valve (level 1.5), she had immediate relief that has persisted at 3 years follow-up.
Case 3 (AH, communicating)
A 71-year-old man who presented to us 20 years after the onset of symptoms, had severe difficulty with gait and balance, thinking and memory, and bladder control. He had a risk factor for AH insofar as when he was a young man in college, he boxed and had been knocked out several times. His gait symptoms began about age 50 years, and involved loss of balance, pitching forward, and shuffling. A CT scan at that time showed ventriculomegaly, but his symptoms were not attributed to this finding, and he was not treated. His cognitive difficulties began ~7 years later, and were manifest by difficulty with calculations and paperwork. Within 10 years, he needed a walker, had significant cognitive decline, and severe urinary urgency and frequency. In the last 4 years prior to presentation, he became wheelchair-bound, incontinent, and had profound cognitive decline to the extent that he was nearly mute. Upon our exam, the head circumference was between the 75th and 90th percentiles, and the Mini Mental State Examination score was 5/30. After careful discussion with his family, including review of the potential risks and uncertainty of benefits, we proceeded with shunt surgery without performing physiologic testing. A Codman-Hakim programmable shunt with an anti-siphon device (Heyer-Schulte) was inserted with an initial setting of 200 mm H2O, and was gradually taken to the lowest setting over the ensuing year, with no improvement in any of his symptoms.
Case 4 (IH, obstructive)
A 53-year-old woman had unsteady gait, headaches, worsening job performance, and recurrent syncope for several months. Her medical history was otherwise unremarkable. She had been referred for outpatient evaluation, but before this occurred, she was found in her home lethargic and complaining of severe headache. Upon arrival to the hospital she was unresponsive, bradycardic, and hypertensive. Neuroimaging revealed massive ventricular dilatation and fourth ventricle outlet obstruction (Figure 2). She was initially treated with external ventricular drainage, and then a PS Medical Delta valve (level 1.5) was inserted. More than 2 years later she has a normal gait, no syncope, no chronic headaches, and she is employed.
The signs and symptoms of hydrocephalus are well known in children4,5,11 and the elderly12-16, but descriptions of its presentation in young and middleaged adults have largely been restricted to obstructive hydrocephalus (LIAS or LOVA)6,7. Because adults in this age range have been included in cohorts of predominately elderly patients with NPH4,15,17 the clinical presentation of young adults has not been differentiated. Our results show that the signs and symptoms of hydrocephalus in the 16-55 year age range are similar regardless of the etiology of the hydrocephalus. Thus, age, rather than etiology, appears to be the main determinant of the clinical syndrome. Therefore, we propose the recognition of a single, clinically distinct syndrome of hydrocephalus in young and middle-aged adults (SHYMA) that, if associated with ventriculomegaly, should prompt further investigation of possible pathophysiology of CSF flow, pressure, or resorption, and consideration of the need for surgical therapy.
In SHYMA, there is a contrast between the severity of patients' complaints and the subtlety of findings on clinical examination (Table 2). In our cohort, the distribution of primary symptoms (gait 70%, cognition 70%, urinary urgency 48%) are similar to those in NPH in the elderly3,12,15, but the subtlety of the symptoms and their differing effect on patients' livelihoods distinguishes them from NPH in the elderly. Gait symptoms consisted of clumsiness, difficulty with uneven surfaces, and difficulty with stairs, yet gait apraxia was notably absent, and
In addition to the three NPH-like primary symptoms, headache was encountered frequently enough to be considered a fourth primary symptom of SHYMA. Headache has been well described in LIAS6 and LOVA7, and it was present in 21 out of 30 (70%) patients from our cohort with obstructive hydrocephalus, but headache has not been previously described in communicating hydrocephalus. In fact, 11 out of 16 (69%) patients from our cohort with communicating hydrocephalus presented with headaches, and eight out of eleven (73%) who were treated, had headache resolution.
In combination, the primary symptoms of SHYMA had a profound effect on patients' lives. Job performance was impaired in 84% of surveyed employed patients, not only by the cognitive changes and headaches, but also by the urinary urgency and frequency that often meant leaving one's desk or job station frequently. Many faced disciplinary action or loss of job as a result.
The discrepancy between prominent symptoms and subtle signs, and the relative infrequency with which most physicians see NPH18 may have made it difficult to recognize symptomatic hydrocephalus in patients of this age range, resulting in delayed diagnosis. The mean length of time-to-diagnosis in this cohort (6 years) suggests that this was, indeed, the case. All patients had sought medical attention, yet symptoms referable to hydrocephalus were attributed to other causes, and often not even considered to be correlated to the ventriculomegaly evident on neuroimaging. In fact, a common response from most of our patients upon diagnosis was their tremendous relief that their symptoms were acknowledged as real, rather than related to job dissatisfaction, mid-life crisis, depression, or other psychiatric problems, as many had been told.
In addition to the four primary symptoms, other symptoms in our cohort were relieved by treatment of hydrocephalus, and thus appear to be supporting symptoms for the diagnosis of SHYMA. These include visual disturbances, change in sleep patterns, and syncope. In particular, the occurrence of syncope, which has rarely been described as a symptom of hydrocephalus in the absence of intraventricular cysts or masses, is worrisome because it could potentially represent either transient impairment of cerebral blood flow resulting from ICP elevation, or transient herniation of the medulla through the foramen magnum. In fact, the patient described as case 4 acutely deteriorated with clinical signs and symptoms of cerebral herniation. We believe that syncope occurring in the presence of ventriculomegaly is a relatively urgent indication to investigate the need for surgical treatment of hydrocephalus.
A particularly important subset of our cohort includes the three patients who had been diagnosed and surgically treated for hydrocephalus in infancy or childhood. As they matured and no longer received their medical care from pediatricians or pédiatrie specialists, they also no longer received attention or follow up for their hydrocephalus. We postulate that their original shunts-including one that was over 40 years old-became obstructed at some unknown time, and that patients remained asymptomatic because of adequate compensatory mechanisms. As these patients grew older, the compensatory mechanisms for hydrocephalus gradually failed, and new symptoms emerged.
The lack of transition of their care to adult specialists may have contributed to the unrecognized development of signs and symptoms of decompensating hydrocephalus.
As a result, we believe that the first generation of infants and children treated with shunts for hydrocephalus, who are now 40-50 years old, and the countless other adults with shunts that were put in while they were children, constitute a large group of patients known to be at risk for incipient decompensation and development of SHYMA. We encourage the creation of systems that support the transition and continuity of care for patients with hydrocephalus from their pediatricians and pediatric specialists to their counterparts in adult medicine, with the aim of preventing, or identifying and treating insidious development of SHYMA. The outcome of treatment in our patients offers hope that late decompensation of previously treated hydrocephalus can be treated successfully, and this issue should be studied prospectively2.
Prompt recognition of symptoms, and accurate diagnosis of SHYMA are crucial, as they can alleviate stress for patients, especially related to job performance or the risk of loss of employment, and may even prevent disability. The patient described in case 3 initially developed symptoms at age 50, yet he did not receive treatment until he was severely impaired at the age of 71. His family was willing to try surgical treatment with the understanding that he may already have irreversible neuronal injury that would preclude his recovery, which regrettably was the outcome 1 year after shunt surgery. This example of a very poor outcome after two decades of untreated symptomatic hydrocephalus is a most important reason to recognize and treat SHYMA.
We believe the progress of SHYMA from subtle physical findings to severe impairment can be prevented. Overall, our patients responded well to surgical intervention, with 93% having at least partial improvement, and 56% having complete recovery. ETV resulted in a sustained reduction in symptoms in only five out of eleven patients, the other six were subsequently treated with VP shunts. This poor outcome with ETV, in comparison with other reports6,19,20, is difficult to explain. It is possible that our patients had both communicating and obstructive hydrocephalus pathophysiology; however, we did not assess CSF outflow resistance. Further investigation of the role of ETV in SHYMA from obstructive hydrocephalus may be warranted. In total, the need for surgical revision in this series was similar to other populations12,15,21, which reflects the limitations of current CSF diversion technology. The long-term prognosis with treatment remains to be determined.
There are limitations to our study. It is retrospective, and the potential for recall bias on the part of patients or their families exists. The duration of symptoms was correlated with medical records whenever possible. The cohort is a convenience sample that was not studied prospectively, so that assessment of the signs and symptoms is general, rather than specific. Nonetheless, there is enough detail and consistency to describe the syndrome, and it should be studied prospectively in more detail.
In summary, we propose there exists a single, clinically distinct syndrome of hydrocephalus in young and middle-aged adults (SHYMA) that comprises hydrocephalus of all etiologies. SHYMA is characterized by four primary symptoms (impaired gait, cognition and bladder control, and headaches), a discrepancy between the prominence of the symptoms and the subtlety of clinical signs, and supportive symptoms of visual disturbances, change in sleep pattern, or syncope. A prominent concern of patients is impairment of job performance resulting from the combined effects of these symptoms. The 1-year outcome of surgical therapy is good. We hope that awareness of SHYMA will increase the likelihood of early diagnosis and treatment, and prevent loss of employment or disability.
Dr Williams and Dr Rigamonti receive financial support for hydrocephalus research from Medtronic. This research is unrelated to the topic of this manuscript. Dr Williams has received honoraria from Medtronic and from Codman to speak about hydrocephalus.
We wish to thank Emily Fudge, Executive Director of the Hydrocephalus Association, for asking Michael A. Williams, MD, to make a presentation on hydrocephalus in young adults at the Hydrocephalus Association meeting in Phoenix, Arizona, May 26-28, 2000. Her request, and the subsequent analysis of our data resulted in this manuscript. Interested health professionals, patients and families can contact: www.hydroassoc. org. We also wish to acknowledge Medtronic, Inc., for their financial support of this project.
1 Casey AT, Kimmings EJ, Kleinlugtebeld AD et al. The long-term outlook for hydrocephalus in childhood. A ten-year cohort study of 155 patients. Pediatr Neurosurg 1997; 27: 63-70
2 Sgouros S, Malluci C, Walsh AR et al. Long-term complications of hydrocephalus. Pediatr Neurosurg 1995; 23: 127-132
3 Adams, RD, Fisher CM, Hakim S et al. Symptomatic occult hydrocephalus with 'normal' cerebrospinal fluid pressure (a treatable syndrome). N Engl J Med 1965; 273: 117-126
4 Bret P, Chazal J. Chronic normal pressure hydrocephalus in childhood and adolescence. A review of 16 cases and reappraisal of the syndrome. Childs Nerv Sysf 1995; 11: 687-691
5 Barnett GH, Hahn JF, Palmer J. Normal pressure hydrocephalus in children and young adults. Neurosurgery 1987; 20: 904-907
6 Fukuhara T, Vorster SJ, Luciano MC. Risk factors for failure of endoscopie third ventriculostomy for obstructive hydrocephalus. Neurosurgery 2000; 46: 1100-1111
7 Oi S, Shimoda M, Shibata M et al. Pathophysiology of long-standing overt ventriculomegaly in adults. J Neurosurg 2000; 92: 933-940
8 Bushby KM, Cole T, Matthews JN et al. Centiles for adult head circumference. Arch Dis CWW 1992; 67: 1286-1287
9 Williams MA, Razumovsky AY, Hanley DF. Comparison of Pcsf monitoring and controlled CSF drainage diagnose normal pressure hydrocephalus. Acfa Neurochir Suppl 1998; 71: 328-330
10 Folstein MF, Folstein SE, McHugh PR. 'Mini-mental state'. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12: 189-198
11 Whittle IR, Johnston IH, Besser M. lntracranial pressure changes in arrested hydrocephalus. J Neurosurg 1985; 62: 77-82
12 Black PM. ldiopathic normal-pressure hydrocephalus. Results of shunting in 62 patients. J Neurosurg 1980; 52: 371-377
13 Caruso R, Cervoni L, Vitale AM et al. Idiopathic normal-pressure hydrocephalus in adults: Result of shunting correlated with clinical findings in 1 8 patients and review of the literature. Neurosurg Rev 1997; 20: 104-107
14 Vanneste J, Augustijn P, Dirven C et al. Shunting normal-pressure hydrocephalus: Do the benefits outweigh the risks? A multicenter study and literature review. Neurology 1992; 42: 54-59
15 Larsson A, Wikkelso C, Billing M et al. Clinical parameters in 74 consecutive patients shunt operated for normal pressure hydrocephalus. Acfa Neurol Scancl 1991; 84: 475-482
16 Raftopoulos C et al. Cognitive recovery in ldiopathic normal pressure hydrocephalus: A prospective study. Neurosurgery 1994; 35: 397-405
17 Vanneste J, Augustijn P, Tan WF et al. Shunting normal pressure hydrocephalus: the predictive value of combined clinical and CT data. J Neurol Neurosurg Psychiat 1993; 56: 251-256
18 Vanneste J, van Acker R. Normal pressure hydrocephalus: Did publications alter management? J Neurol Neurosurg Psychiat 1990; 53: 564-568
19 Mitchell P, Mathew B. Third ventriculostomy in normal pressure hydrocephalus. Br J Neurosurg 1999; 13: 382-385
20 Tisell M, Almstrom O, Stephensen H ef al. How effective is endoscopie third ventriculostomy in treating adult hydrocephalus caused by primary aqueductal stenosis? Neurosurgery 2000; 46: 104-111
21 Zemack G, Romner B. Seven years of clinical experience with the programmable Codman Hakim valve: A retrospective study of 583 patients. J Neurosurg 2000; 92: 941-948
John A. Cowan[dagger][double dagger], Matthew J. McGirt[dagger], Graeme Woodworth[dagger], Daniele Rigamonti[dagger] and Michael A. Williams[dagger]*
Johns Hopkins Medical Institutions Adult Hydrocephalus Program, Departments of * Neurology and [dagger] Neurosurgery, Baltimore, MD, USA
[double dagger] Now affiliation with: Department of Neurosurgery, University of Michigan, Ann Arbor, MI, USA
Correspondence and reprint requests to: Michael A. Williams, MD, Johns Hopkins Hospital Adult Hydrocephalus Program, 600 N. Wolfe Street, Phipps 102, Baltimore, MD 21287, USA. [mwilliam@jhmi. edu] Accepted for publication October 2004.
Copyright Maney Publishing Jul 2005
Provided by ProQuest Information and Learning Company. All rights Reserved