Epidemiological and Clinical Features of Moyamoya Disease in the USAKainth D. · Chaudhry S.A. · Kainth H. · Suri F.K. · Qureshi A.I.
Zeenat Qureshi Stroke Research Center, Departments of Neurology, Neurosurgery, and Radiology, University of Minnesota, Minneapolis, Minn., USA Corresponding Author
Background: An increasing number of cases of Moyamoya disease have been reported in the Japanese and US literature. We performed this study to quantify the rise in the prevalence of Moyamoya disease and to study the unique epidemiological and clinical features in the USA that may explain a change in incidence. Methods: We analyzed data derived from patients entered in the Nationwide Inpatient Sample between 2005 and 2008, using ICD-9 codes for Moyamoya disease. Data including patient age, gender, ethnicity, secondary diagnosis, medical complications, and hospital costs were obtained. Results: From 2005 to 2008 in the USA, there were an estimated 7,473 patients admitted with a primary or secondary diagnosis of Moyamoya disease. Patients admitted with Moyamoya disease were most frequently women and Caucasian. Overall, ischemic stroke was the most common reason for admission. Hemorrhagic stroke was more frequent in adults compared with children, 18.1 versus 1.5% (p < 0.05). Conclusion: The number of patients identified and admitted with Moyamoya disease has risen dramatically in the last decade. This study can lead to a better understanding of the disease pattern and healthcare consequences in the USA and suggests that pathophysiologic differences in Moyamoya disease may exist.
Copyright © 2013 S. Karger AG, Basel
Moyamoya disease is a chronic cerebral occlusive vasculopathy resulting in progressive narrowing of the major intracranial vessels . Subsequently, occlusion of the supraclinoid internal carotid arteries and of the circle of Willis occurs, accompanied by proliferation of arterial collaterals in proximity to the site of occlusion at the base of the brain. Moyamoya, which translated means ‘puff of smoke', was the term coined by Suzuki and Takaku in 1969 to describe the angiographic appearance of the network of collaterals . Our understanding regarding the clinical and epidemiological features of Moyamoya disease is predominantly derived from studies conducted in the Japanese population [3,4,5]. However, there is limited data regarding the clinical and epidemiological features of Moyamoya disease in populations such as the USA, which is comprised of various ethnicities. Small single-center studies in the USA have demonstrated that features of Moyamoya disease may be different from those observed in Japan [6,7]. The Japanese literature has reported that the prevalence of Moyamoya disease has doubled from 3,900 in 1994 to 7,700 in 2003 . This increasing incidence is also suggested by case series in the USA [7,8]. Further study may lead to a better understanding and awareness of the disease pattern in the USA and potentially identify factors responsible for the rising incidence of Moyamoya disease. We performed this study to identify the clinical and epidemiological features of Moyamoya disease using nationally representative data from the USA.
Materials and Methods
The data were obtained from the Nationwide Inpatient Sample (NIS). The NIS is sponsored by the Agency for Healthcare Research and Quality, and data are collected from roughly 1,000 hospitals of 7-8 million admissions annually. These data provide a 20% stratified sample of acute care hospital admissions which can be extrapolated for deriving nationwide estimates. Different geographic areas of the USA are represented, including the Northeast, Midwest, West, and South. Detailed information about the design of the NIS is available at http://www.hcup-us.ahrq.gov. The NIS includes >100 clinical and nonclinical variables for each hospital stay. The variables include primary and secondary diagnoses, primary and secondary procedures, patient demographics (including age, gender, race), admission and discharge information, and length of stay. This paper contains a statistical analysis of data obtained from the NIS.
Patients registered in the NIS from 2005 to 2008 were included in the analysis. Patients with Moyamoya disease were identified using the International Classification of Disease, 9th revision, clinical modifier (ICD-9-CM) code 7437.5. We performed a hospital-based study (of two hospitals) to validate the methodology of ascertainment of Moyamoya disease patients in the NIS database based on the accuracy of ICD-9-CM code 7437.5. We identified all patients discharged with the primary or secondary ICD-9-CM code 7437.5 between 2008 and 2010 at the two university-affiliated tertiary care hospitals in Minneapolis, Minn., and verified the accuracy by concurrent chart review. We found that the ICD-9 code had a positive predictive value of 89% among 27 patients identified.
ICD-9-CM secondary diagnosis codes were used to identify associated complications such as ischemic stroke (433, 434, 436, 437.0, 437.1), transient ischemic attack (435.9), hemorrhagic stroke (430, 431, 432), requirement of mechanical ventilation (96.72) or blood transfusion (99.04). The ICD-9-CM procedure codes were also used to evaluate the proportion of patients with Moyamoya disease who underwent associated procedures such as cerebral angiography (88.41) or intracranial angioplasty/stent placement (00.62). Age, gender, race/ethnicity, associated diagnoses, procedures performed, length of stay, discharge status, and total hospitalization charges were determined for all cases. We categorized routine discharge as minimal disability, and any other discharge status as moderate to severe disability.
The software program SAS 9.1 (SAS Institute Inc., Cary, N.C., USA) was used to convert raw counts generated from the NIS database into weighted counts to provide national estimates, following recommendations of the Healthcare Cost and Utilization Project. . We used univariate analysis, the χ2 test for categorical data, and the t test for continuous data to identify differences in study variables and end points between patient groups. We categorized patients based on age at time of admission (<18, 19-45, and >45 years) since Moyamoya disease has different clinical presentations based on age.
From 2005 to 2008, there were an estimated 7,473 patients admitted with a primary diagnosis of Moyamoya disease in the USA. Table 1 summarizes the inpatient demographic and clinical data of these patients. There were an estimated 2,236 pediatric admissions and 5,237 adult admissions. Women comprised the majority of the admissions (69.1%). Patients admitted with Moyamoya disease were most frequently Caucasian. Overall, ischemic stroke was the most common reason for admission among both children and adults. Hemorrhagic stroke was more frequent in adults compared with children, (18.1 vs. 1.5%, p < 0.05). There were no differences seen in the distribution of presentations such as transient ischemic attack, ischemic stroke, or hemorrhagic stroke seen among strata defined by race/ethnicity. There were a high proportion of patients with hypertension and diabetes mellitus among patients with Moyamoya disease, 36.5 and 15.9%, respectively.
|Table 1. Demographic and clinical characteristics and in-hospital outcomes within age strata among patients admitted with primary diagnosis of Moyamoya disease in the US (NIS 2005-2008)|
Cerebral angiography was the most common procedure performed during hospitalization (in 24% of the patients) in patients with Moyamoya disease. The frequency of cerebral angiography was higher in adults compared with children, (27 vs. 17%, p < 0.05). Children required blood transfusions more frequently during their admission compared with adults (11.9 vs. 6.6%, p < 0.05).
The average length of stay was similar between children and adults, 5.9 versus 6.7 days. Of all the inpatients, 2.2% died during their hospital admission. The overall estimated hospitalization charges were higher in adults aged 19-45 compared with children, USD 57,000 versus USD 44,000. Similarly, the hospitalization charges in the age group >45 years were even higher, estimated at USD 68,000 (table 2).
|Table 2. Demographic and clinical characteristics and in-hospital outcomes within strata defined by hospital presentation among patients admitted with primary diagnosis of Moyamoya disease in the US (NIS 2005-2008)|
Patients admitted with hemorrhagic stroke were more likely to have a longer hospital stay, an average of 15.9 days compared with 6.6 days for patients admitted with ischemic stroke. In addition, patients admitted with hemorrhagic stroke had higher hospitalization costs and more frequently underwent ventriculostomy and required placement of a ventriculoperitoneal shunt. A small number of patients had placement of an intracranial stent or underwent a cranial operation during their admission.
Figure 1 demonstrates the age at admission along with presenting symptoms. Peaks were witnessed in the first and fourth decades of life corresponding to admission for ischemic stroke. In terms of admission associated with hemorrhagic stroke, peaks were seen in the sixth and seventh decades of life (fig. 2).
|Fig. 1. Age distribution of patients admitted with primary diagnosis of Moyamoya disease and ischemic stroke in the USA (NIS 2005-2008).|
|Fig. 2. Age distribution of patients admitted with primary diagnosis of Moyamoya disease and hemorrhagic stroke in the USA (NIS 2005-2008).|
While the clinical and angiographic features of Moyamoya disease are described in previous studies [10,11], the etiology remains unknown. However, Moyamoya syndrome has been associated with diverse pathological conditions such as Down's syndrome, neurofibromatosis, sickle cell disease, trauma, radiation exposure, and congenital giant cervicofacial hemangiomas [6,12,13]. There is evidence for a genetic and familial attribute to the disease and possibly an inflammatory component [14,15].
Moyamoya disease is most prevalent in Japan (6.03 per 100,000) and the prevalence in the USA is thought to be much lower . However, Moyamoya is no longer thought of as a disease solely of Asian populations as an increasing occurrence in non-Asian populations has been demonstrated. Over 6,000 cases have been reported throughout the world . Caucasians represented the majority of cases in the USA; however, they also represent 72.4% of the US population . Overall, the disease appears to be distributed among the races according to their relative proportions of the US population. On the other hand, women are more frequently affected than men. The cause of this is unknown but may be related to underlying genetic or hormonal differences. Moyamoya demonstrates a bimodal age distribution with peaks in the first and fourth decades of life  according to both the US and Japanese literature [2,8,18]. Interestingly, the familial occurrence of Moyamoya disease is less common in the USA (approx. 2%), in contrast to Japan where a family history of Moyamoya disease is present in 12.1% of patients . This finding suggests that perhaps environmental factors play a greater role in the development of Moyamoya disease in the USA.
An increasing number of cases of Moyamoya disease are being reported. Prior to 1997, fewer than 100 cases of the disease had been reported in the USA . Subsequent case series and single-center data reports have demonstrated that the number of Moyamoya cases continues to rise. From 1988 to 2004, roughly 2,247 hospital admissions for Moyamoya disease were recorded . In our study from 2005 to 2008, 7,473 patients were admitted with the disease in the USA. Thus, there has been approximately a 4-fold increase in the prevalence of Moyamoya disease in the USA. The rising trend may be the result of improved disease awareness and diagnosis, but may also represent a growth in the disease prevalence. Similarly, the Japanese literature has also reported a rise in the prevalence of Moyamoya disease. The estimated prevalence doubled in Japan from 3,900 to 7,700 patients from 1994 to 2003 . The prevalence of Moyamoya disease appears to be increasing worldwide and this trend must be closely monitored and investigated to identify potential associated risk factors.
According to our report, ischemic stroke remains the most common presentation in the USA in all age groups. In Japan and Korea, roughly 80% of the children with Moyamoya disease present with an ischemic stroke. Likewise, the majority of the children in our study were also admitted with ischemic Moyamoya disease. Of the adults admitted with symptomatic Moyamoya disease, 33.6% presented with hemorrhagic Moyamoya disease. Higher rates of hemorrhagic Moyamoya disease in adults are witnessed in Japan and Korea, 51 and 69%, respectively . The lower incidence of hemorrhagic Moyamoya disease in adults is also seen in other US studies [6,8,20]. The underlying pathophysiology of Moyamoya disease in the USA may have distinct features that account for the lower hemorrhagic rate in adults. Moyamoya disease may have a lower rate of progression and/or clinical events in the US population compared with Asian populations. The lower rate of progression may allow for more time for developing appropriate collaterals and thus account for the lower rate of hemorrhage in the US population. The disease is characterized by intimal thickening of the walls of the terminal portions of the carotid arteries bilaterally and proximal anterior and middle cerebral arteries leading to the formation of fragile collaterals. The fragile Moyamoya vessels that form at the base of the brain are thin walled and dilated or thick walled and stenotic and are prone to hemorrhage . In addition, underlying pathological differences may exist in the fragility or development of collaterals that may account for the difference in the rate of hemorrhage between the USA and Japan.
There are limitations that should be considered prior to interpretation of our results. There are shortcomings of the NIS data in terms of quality, detail and reliability. For example, the ICD-9 code 7437.5 is the code for Moyamoya disease and does not differentiate between cases of Moyamoya disease and Moyamoya syndrome. This limitation in the code prevents determining whether the increase in cases of Moyamoya diagnoses is due to an increase in Moyamoya disease or in Moyamoya syndrome. In addition, the increase may be due to improved disease awareness and diagnosis or improved coding practices.
Differences in coding practices at various hospitals may affect the quality and reliability of the data. Furthermore, since the NIS only collects data from inpatient admissions, there are incomplete details on the full population as outpatient data is not included.
The prevalence of Moyamoya disease in the USA appears to be increasing. The low familial occurrence of Moyamoya disease in the USA suggests that environmental factors may be contributing to the disease. As we move forward, it will be important to further investigate the modifiable risk factors that can reduce the incidence of this progressive neurological disease.
Daraspreet Kainth, MD
University of Minnesota
Mayo Memorial Building D429, 420 SE Delaware Street MMC 96
Minneapolis, MN 55455-0374 (USA)
Received: June 15, 2012
Accepted: November 19, 2012
Published online: February 23, 2013
Number of Print Pages : 6
Number of Figures : 2, Number of Tables : 2, Number of References : 21
Vol. 40, No. 4, Year 2013 (Cover Date: May 2013)
Journal Editor: Feigin V.L. (Auckland)
ISSN: 0251-5350 (Print), eISSN: 1423-0208 (Online)
For additional information: http://www.karger.com/NED