A Nationwide Analysis of Budd–Chiari Syndrome in the United States

Abstract

Objective

The Budd–Chiari Syndrome (BCS) is a rare disorder characterized by hepatic venous outflow obstruction. The primary objectives of our study were to assess temporal trends in the prevalence of BCS among hospitalized patients in the United States using the National Inpatient Sample (NIS) database and to evaluate demographics, risk factors, and common presentation of BCS.

Methods

Data were extracted from the NIS to identify patients >18 years of age using all listed diagnosis of BCS from 1998 to 2017 and analyzed.

Results

Between 1998 and 2017, we identified a total of 8435 hospitalizations related to BCS. Over the 19-year period, the hospitalization rate for BCS increased consistently from 4.96 per 1,000,000 US population in 1998 to 10.44 per 1,000,000 in 2017, with an annual percentage change increase of 4.41% (95% confidence interval [CI]: 4.23%–4.59%, P < 0.0001). The most common risk factor (7.75%) was myeloproliferative disorder (essential thrombocythemia, polycythemia vera, myelofibrosis, chronic myeloid leukemia) followed (7.32%) by a hypercoagulable state (primary thrombophilia, protein C deficiency, factor V Leiden mutation, antiphospholipid antibody syndrome or prothrombin gene mutation) and paroxysmal nocturnal hemoglobinuria (1.63%). Cirrhosis was present in 18.7%, Portal vein thrombosis in 7.9%, and inferior vena cava thrombosis in 6.4%. The most common manifestations of BCS were ascites (29.9%) or acute kidney injury (18.8%) followed by hepatic encephalopathy (9.6%) and acute liver failure (5.6%).

Conclusion

This large population-based study from the United States showed increasing hospitalizations related to BCS. Common presentation was ascites and acute kidney injury.

The Budd–Chiari Syndrome (BCS) is a rare and potentially lethal disease characterized by hepatic venous outflow obstruction resulting in hepatic congestion and portal hypertension. The outflow obstruction, partial or complete, may occur in small or large hepatic veins, the suprahepatic segment of the inferior vena cava (IVC), or the right atrium. The clinical spectrum of BCS is highly variable and may manifest either as fulminant or acute liver failure to subacute, chronic, or asymptomatic forms. Important clinical features include abdominal pain, ascites, hepatosplenomegaly, and prominent venous collaterals on the trunk and flanks secondary to IVC obstruction.

There are wide geographical variations in the epidemiology and pathogenesis of this disease. In Europe, the prevalence of BCS is estimated to be approximately 1 per 2.5 million population per year, where as in Nepal and India the prevalence is much higher.¹ Similarly, while prothrombotic risk factors are more prevalent among patients with BCS in the Western countries, membranous obstruction either in the IVC and/or hepatic veins is more common in South Asia.² The current knowledge about the epidemiology of BCS in the United States is limited and is based on data from case series and multicenter studies that consisted of a relatively small number of patients.3, 4, 5 The primary objective of our study was to assess the temporal trends in the prevalence of BCS among hospitalized patients in the US using the National Inpatient Sample (NIS) database and also to evaluate demographics, risk factors, and common presentation of BCS.

Patients and methods

Data source

In this retrospective study, we extracted data from the NIS from 1998 to 2017. The NIS is the largest publicly available all-payer inpatient administrative database developed by the Agency for Healthcare Research and Quality for the Healthcare Cost and Utilization Project (HCUP). It represents approximately 20% stratified sample of discharges from community hospitals but excludes long-term acute care hospitals and rehabilitation facilities. The database contains information of more than 7 million hospital discharges annually. The number of states participating in the NIS grew from 8 in 1988 to 46 in 2016. The database captures information about primary and secondary diagnoses during each hospital stay as well as information about procedures. The NIS also contains other valuable information such as severity and comorbidity measures, hospital characteristics (size, region, bed size, teaching/nonteaching), payment source (Medicare/Medicaid/private), total charges, and length of hospital stay. In 2012, the NIS revised the sample design so as to represent a sample of discharges rather than a sample of hospitals. This new strategy is expected to make the estimates more precise by reducing the sampling error. Starting October 1, 2015, all hospitals in the United States adopted International Classification of Diseases (ICD) 10 codes for disease classification and for procedures. The calendar years of 2016 and 2017 which were included in this study used ICD 10 CM/PCS codes.

Population

Data were extracted from the NIS to identify patients >18 years of age using all listed diagnosis (primary or secondary diagnosis) of BCS from 1998 to 2017. The diagnosis of BCS was captured using the codes 453.0 (ICD 9) and I82.0 (ICD 10).

Clinical characteristics

We obtained information on patient (age, sex, race) and hospital characteristics (region of the country, bed size, teaching status) and insurance status (Medicare, Medicaid and private insurance). We identified underlying risk factors for BCS such as myeloproliferative disorders, paroxysmal nocturnal hemoglobinuria, or primary hypercoagulable conditions. We also recorded potential complications including acute liver failure, acute kidney injury, cirrhosis, manifestations of portal hypertension such as ascites, hepatic encephalopathy, and esophageal varices; with and without bleeding, portal vein thrombosis, and spontaneous bacterial peritonitis. Severity of illness was measured using the Elixhauser comorbidity index after excluding liver diseases, and this included 29 major Elixhauser comorbidity conditions.⁶ (Supplementary table 1 shows full list of ICD 9 and ICD 10 codes associated with specific diagnosis and procedures).

Statistical analysis

Descriptive statistics was used to summarize patients’ characteristics, hospital characteristics and utilization, comorbidities, complications, procedures, and the outcomes by using the weighted survey methods. Data were presented as mean and standard error for continuous variables and percentage and standard error for categorical variables. Standard errors of percentage or mean were estimated using the Taylor series linearization method. To make inferences regarding the national estimates for the total number of BCS discharges across the study period, sample weights were applied to each admission per recommendations from the NIS. To adjust for changes to the NIS design in 1998 and 2012, for years before 2012, the trend weight (TRENDWT) was used in place of the original discharge weight (DISCWT) to create national estimates for trend analysis that are consistent with 2012 data onward.⁷ We calculated BCS discharges rate per 1,000,000 US populations by dividing the estimated total BCS discharges by projected US population from the Census Bureau.

The annual percentage change (APC) was derived to compare the patients' characteristics, hospitals’ characteristics, and outcomes over time by using Poisson regression for categorical variables and linear regression with natural logarithm transformation for continuous variables. The p value for APC was used to determine if the trends in the annual percentage change was significantly different from zero; the change was considered as statistically significant with the p value of 0.05 or less. All analyses were performed with SAS version 9.4 (SAS Institute, Cary, NC, USA).

Results

Rate of hospitalization

Between 1998 and 2017, we identified a total of 8435 hospitalizations related to BCS. Using the sample weights provided by HCUP, this corresponds to a total of 41,119 hospital discharges. Over the 19-year period, the hospitalization rate for BCS increased consistently from 4.96 per 1,000,000 US population in 1998 to 10.44 per 1,000,000 in 2017, with an APC increase of 4.41% (CI: 4.23%–4.59%, P < 0.0001) (Figure 1). A majority of these admissions occurred in large hospitals (71.3%) and teaching hospitals (68.9%). During the study period admissions to rural and nonteaching hospitals declined significantly while admissions to urban teaching hospitals increased (APC 1.33%, CI: 0.86%, 1.80%, P < 0.0001). The Southern states contributed to a majority of the BCS admissions (33.5%).

Figure 1.

Trends in hospitalization rates with Budd–Chiari syndrome from 1998 to 2016.

Patient demographics

The mean age of the cohort was 51 years, 55.2% were women and 56.0% were white. The women were younger than men (49.1 ± 0.27 years vs. 52.2 ± 0.28 years, P < 0.0001). While the hospitalization rate for whites marginally increased during this time period, there was a notable and significant increase in the admission rate among Hispanics (4.00%, CI: 2.7–5.3, P < 0.0001) and blacks (2.4%, CI: 1.4–3.5, P < 0.0001). Nearly half (51.6%) the admitted the patients were covered by Medicare/Medicaid (Table 1).

Table 1.

Characteristics of Patients and Hospitals.

Characteristics	1998–2017 (unweighted: 8435 weighted: 41,119)	1998 (unweighted: 262 weighted: 1367)	2017 (unweighted: 680 weighted: 3400)	APC (CI)	P value for APC
BCS patients' characteristics:
Age	50.50 (0.19)	52.01 (1.06)	51.10 (0.65)	0.13% (−0.08%,0.34%)	0.221
Female	55.19 (0.54)	51.71 (3.12)	52.65 (1.92)	−0.43% (−0.93%,0.08%)	0.095
Race
1: White	56.03 (0.54)	54.19 (3.12)	58.24 (1.89)	0.98% (0.46%,1.49%)	<0.0001
2: Black	13.26 (0.37)	15.34 (2.3)	15.29 (1.38)	2.44% (1.35%,3.54%)	<0.0001
3: Hispanic	9.56 (0.32)	6.06 (1.44)	12.21 (1.26)	3.99% (2.66%,5.33%)	<0.0001
4: Asian/Pacific islander	2.56 (0.17)	2.82 (0.99)	2.94 (0.65)	2.09% (−0.36%,4.60%)	0.095
6: Other	3.65 (0.2)	1.92 (0.85)	6.18 (0.92)	6.08% (3.82%,8.40%)	<0.0001
9: Unknown	14.93 (0.39)	19.68 (2.53)	5.15 (0.85)	−8.93% (−9.80%,-8.04%)	<0.0001
Primary payer:
1: Medicare	33.17 (0.52)	33.63 (2.98)	34.85 (1.83)	0.76% (0.09%,1.43%)	0.025
2: Medicaid	18.42 (0.42)	12.77 (2.07)	21.32 (1.57)	2.09% (1.18%,3.02%)	<0.0001
3: Private insurance	40.20 (0.54)	45.73 (3.13)	35.44 (1.84)	−1.58% (−2.16%,-1.00%)	<0.0001
6: Other	8.21 (0.3)	7.88 (1.67)	8.38 (1.06)	0.32% (−1.01%,1.66%)	0.641
Hospital characteristics
Hospital size
1: Small	9.81 (0.32)	7.18 (1.49)	13.97 (1.33)	3.70% (2.41%,5.00%)	<0.0001
2: Medium	18.92 (0.43)	13.63 (2.04)	23.68 (1.63)	1.64% (0.75%,2.55%)	<0.0001
3: Large	71.27 (0.49)	79.19 (2.41)	62.35 (1.86)	−0.91% (−1.35%,-0.46%)	<0.0001
Hospital location and teaching status
1: Rural	6.89 (0.28)	8.35 (1.69)	4.41 (0.79)	−2.63% (−4.00%,-1.25%)	<0.0001
2: Urban nonteaching	24.24 (0.47)	21.52 (2.47)	15.15 (1.38)	−2.77% (−3.49%,-2.03%)	<0.0001
3: Urban teaching	68.87 (0.51)	70.13 (2.79)	80.44 (1.52)	1.33% (0.86%,1.80%)	<0.0001
Hospital region
1: Northeast	21.84 (0.45)	18.29 (2.4)	21.76 (1.58)	0.28% (−0.54%,1.10%)	0.508
2: Midwest	22.15 (0.46)	30.90 (2.98)	19.41 (1.52)	−0.14% (−0.95%,0.67%)	0.736
3: South	33.45 (0.52)	31.19 (2.85)	35.59 (1.84)	0.38% (−0.27%,1.04%)	0.252
4: West	22.57 (0.46)	19.62 (2.4)	23.24 (1.62)	−0.69% (−1.47%,0.10%)	0.087

APC, Aannual Ppercentage Cchange; CI, 95% confidence intervals.

All data are presented as percentage (standard error) for categorical variables and mean (standard error) for continuous variables.

APC >0 means increasing, <0 means decreasing.

P value for APC measures if APC is significantly different from zero. P value ≤ 0.05 means the change is significant.

Risk factors for BCS

The most common risk factor (7.8%) was myeloproliferative disorder (essential thrombocythemia, polycythemia vera, myelofibrosis, chronic myeloid leukemia) followed (7.3%) by a hypercoagulable state (primary thrombophilia, protein C deficiency, factor V Leiden mutation, antiphospholipid antibody [APLA] syndrome or prothrombin gene mutation) and paroxysmal nocturnal hemoglobinuria (PNH, 1.6%) (Table 2).

Table 2.

BCS Risk Factors and Clinical Characteristics.

Risk factors and complications	1998–2017 (unweighted: 8435 weighted: 41,119)	1998 (unweighted: 262 weighted: 1367)	2017 (unweighted: 680 weighted: 3400)	APC (CI)	P value for APC
1. Risk factors
Myeloproliferative disorder	7.75 (0.29)	11.41 (2.01)	6.18 (0.92)	0.11% (−1.25%,1.48%)	0.88
Primary hypercoagulable state	7.32 (0.29)	N/A	11.62 (1.23)	11.43% (9.55%, 13.34%)	<0.0001
PNH	1.63 (0.14)	1.87 (0.83)	0.59 (0.29)	−5.79% (−8.47%,-3.03%)	<0.0001
2. Clinical Characteristics
Ascites	29.93 (0.5)	37.85 (3.02)	26.62 (1.7)	−0.65% (−1.33%,0.03%)	0.06
Acute kidney injury	18.84 (0.43)	9.47 (1.89)	27.35 (1.71)	5.65% (4.66%,6.66%)	<0.0001
Cirrhosis	18.65 (0.43)	12.66 (2.05)	25.00 (1.66)	3.23% (2.29%,4.18%)	<0.0001
Portal hypertension	16.57 (0.41)	13.36 (2.07)	21.18 (1.57)	3.48% (2.48%,4.49%)	<0.0001
Hepatic encephalopathy	9.59 (0.32)	9.00 (1.77)	12.65 (1.28)	2.60% (1.32%,3.91%)	<0.0001
Portal vein thrombosis	7.92 (0.3)	6.33 (1.55)	24.12 (1.64)	10.49% (8.73%,12.27%)	<0.0001
Esophageal varices without bleeding	7.44 (0.29)	5.44 (1.43)	9.56 (1.13)	4.94% (3.38%,6.52%)	<0.0001
Any history of thrombosis/embolism	7.08 (0.28)	0.39 (0.39)	10.88 (1.2)	11.89% (9.96%,13.87%)	<0.0001
Acute respiratory Failure	7.03 (0.28)	3.23 (1.14)	10.88 (1.2)	5.08% (3.48%,6.69%)	<0.0001
HCC	6.93 (0.28)	3.80 (1.18)	8.38 (1.06)	5.40% (3.78%, 7.05%)	<0.0001
Acute blood loss anemia/hemorrhagic	6.62 (0.27)	7.20 (1.61)	10.44 (1.17)	8.00% (6.20%,9.82%)	<0.0001
IVC thrombosis	6.39 (0.27)	5.04 (1.37)	5.29 (0.86)	2.89% (1.30%,4.50%)	0.000
Alcoholic cirrhosis	5.73 (0.25)	7.76 (1.65)	0.74 (0.33)	−3.46% (−4.94%,-1.95%)	<0.0001
Acute liver failure	5.60 (0.25)	4.41 (1.31)	7.79 (1.03)	2.18% (0.52%,3.87%)	0.01
Coagulation disorder	5.25 (0.24)	5.70 (1.44)	5.44 (0.87)	−0.04% (−1.67%,1.62%)	0.96
Pulmonary embolism	3.85 (0.21)	2.57 (0.96)	5.00 (0.84)	3.47% (1.40%,5.58%)	0.001
Hepatorenal syndrome	3.29 (0.2)	4.67 (1.33)	3.38 (0.69)	1.95% (−0.21%,4.16%)	0.078
Variceal bleeding	3.20 (0.19)	5.61 (1.42)	2.35 (0.58)	−4.25% (−6.21%,-2.25%)	<0.0001
Spontaneous bacterial peritonitis	2.83 (0.18)	3.47 (1.14)	3.53 (0.71)	−0.42% (−2.63%,1.84%)	0.71
Liver abscess	2.79 (0.18)	1.03 (0.6)	6.47 (0.94)	11.87% (8.80%,15.02%)	<0.0001
Complications of vascular device/graft	2.74 (0.18)	4.29 (1.33)	2.35 (0.58)	−2.77% (−4.93%,-0.56%)	0.01
Acute pancreatitis	2.50 (0.17)	0.33 (0.33)	2.06 (0.54)	1.55% (−0.88%,4.04%)	0.21
Intestinal infarct/acute vascular insufficiency	2.11 (0.16)	1.82 (0.82)	0.59 (0.29)	−1.94% (−4.40%,0.58%)	0.13
Liver transplant complications	2.09 (0.16)	0.69 (0.49)	1.18 (0.41)	−0.25% (−2.81%,2.37%)	0.85
Elixhauser Comorbidity Score excluding liver disease	9.38 (0.12)	9.02 (0.62)	12.52 (0.48)	2.22% (1.46%,2.98%)	<0.0001

APC, annual percentage change; CI, 95% confidence interval; HCC, hepatocellular carcinoma; IVC, inferior vena cava; PNH, paroxysmal nocturnal hemoglobinuria.

All data are presented as percentage (standard error) for categorical variables and mean (standard error) for continuous variables.

APC >0 means increasing, <0 means decreasing.

P value for APC measures if APC is significantly different from zero. P value ≤ 0.05 means the change is significant.

Clinical presentation

The most common manifestations of BCS among admitted patients were ascites (29.9%), liver cirrhosis (18.65%), and acute kidney injury (18.8%). Other complications of BCS included hepatic encephalopathy (9.6%), acute liver failure (5.6%), esophageal varices without bleeding (7.4%), portal vein thrombosis (7.9%), IVC thrombosis (6.4%), esophageal varices with bleeding (3.2%), and hepatorenal syndrome (3.3%) (Table 2).

Cirrhosis was present in 18.7% of patients, and 6.9% (n = 581) had evidence of hepatocellular cancer (HCC) suggesting that some of these patients had chronic BCS, but a significant proportion had other risk factors. Of the 581 patients with HCC, 36.7% had hepatitis C virus, 14.8% had alcohol related liver disease, and 6% had hepatitis B virus. While the prevalence of variceal bleeding decreased with time (APC: − 4.3%, CI: − 6.2, −2.3, P < 0.0001), the prevalence of other complications of BCS increased from 1998 to 2017, with the most notable increase for portal vein thrombosis (APC: 10.5%, CI: 8.7–12.3, P < 0.0001), HCC (APC: 5.40%, CI: 3.8–7.1, P < 0.0001), and acute kidney injury (APC: 5.7%, CI: 4.7–6.7, P < 0.0001.

Discussion

This is the largest population-based study from the United States that analyzed hospitalizations related to BCS with a high degree of granularity. During the study, there has been a significant increase in the number of BCS admissions from 4.96 per 1 million in 1998 to 10.44 per 1 million in 2017 corresponding to an annual percentage change (APC) increase of 4.41% (P < 0.0001). While there were only 1,367 hospitalizations (weighted number) for BCS in 1998, there were 3400 hospitalizations in 2017. This may be a reflection of an overall increase in the incidence and prevalence of patients with BCS or possibly related to better imaging modalities that is capable of detecting hepatic venous outflow tract obstruction early in the disease process with a high degree of precision and accuracy. Reporting bias or awareness of BCS may not explain this increase in BCS. Another possibility is that patients with BCS are living longer and we are capturing these patients when they are admitted with complications as shown by increasing prevalence of cirrhosis and HCC.

There is a paucity of data regarding the prevalence of BCS in the USA. Most of our current knowledge regarding this rare disorder is based on epidemiological studies done in Europe and Asia.^2,8 According to a recent study from France involving 48 centers, the incidence and prevalence of BCS in 2010 were 0.45 and 2.87 cases per million people respectively.⁹ An epidemiological survey conducted in Japan in 1989 concluded that the prevalence of BCS was approximately 2.4 per million with 20 new cases occurring each year.¹⁰ In Nepal and India, the prevalence is thought to be much higher than the West, although the exact prevalence remains unknown.² In our study, we found out that the nationwide inpatient prevalence of BCS in 2017 was 10.44 per million US population.

In Western countries, BCS was thought to be more prevalent among women whereas in Asia, the distribution is considered to be fairly equal between men and women. The higher prevalence in Western countries is often attributed to the increased use of oral contraception.² A single center study of 47 patients with BCS from the USA showed that 66% were women, and similarly a study from France reported that 70% of 173 patients were women.^5,9 A national registry report from the USA that analyzed 510 liver transplants for BCS also had a majority (67%) of women.¹¹ In our current study, there was a higher proportion (55%) of women, but the difference was significantly lower than previously reported. We believe that our estimates are likely to be more accurate than previous estimates using small sample size.

In our study, we found that the prevalence of myeloproliferative disorders among hospitalized patients with BCS was 7.8%, while 7.3% had a primary hypercoagulable state as an underlying etiology. These numbers are significantly lower than previously reported and could be due to inadequate diagnostic work up that was performed during their inpatient stay. The wide array of biochemical tests that can help identify risk factors for BCS are often performed in the outpatient setting and could perhaps explain some of our findings. It is also worth pointing out that in the acutely-ill hospitalized patient with hepatic vein thrombosis and liver dysfunction, the plasma levels of clotting factors are often unreliable and may have dissuaded physicians from investigating them for clotting abnormalities. Some of the laboratory reports may not be also available at the time of hospital discharge. Previous studies had suggested that approximately half of the patients with BCS have some form of myeloproliferative neoplasm, the most common one being the polycythemia vera (10%–40%).^1,5 Inherited thrombophilias such as factor V Leiden mutation, protein C deficiency, protein S deficiency, and APLA syndrome are also important predisposing factors, and of these factor V Leiden mutation is seen in 12%–25% of patients.^12,13

The heterogenous clinical presentation of BCS is well-documented and can range from fulminant liver failure to an asymptomatic presentation.¹⁴ The clinical manifestations of BCS depend on the site of obstruction (pure hepatic vein thrombosis vs. isolated IVC thrombosis or a combination of both) and the pace at which obstruction of the hepatic venous outflow tract occurs. In our cohort of hospitalized patients with BCS, 9.6% had hepatic encephalopathy (HE) and 3.2% had variceal bleeding. It is plausible that this subgroup (~13%) of patients had a rather fulminant or acute form of BCS that required hospitalization. BCS manifesting as fulminant liver failure is relatively uncommon, and according to one multicenter European study that consisted of 163 patients, 9% had HE at the time of BCS diagnosis.¹²

Ascites is considered the most common manifestation of BCS, and it has been suggested that 70–90% of patients with BCS will develop ascites.^12,15,16 In our hospitalized patients with BCS, however, only 30% patients had ascites. A study from Japan had reported ascites only in 31% of their patients, but attributed this to the fact that a majority of their patients had membranous occlusion of the IVC (93%) with patent hepatic veins.¹⁰ In our study, only 6.8% had concomitant IVC thrombosis along with hepatic vein thrombosis. The incidence of spontaneous bacterial peritonitis (SBP) in BCS is unknown, but 2.8% of the hospitalized patients in our cohort had SBP. The lower prevalence of ascites in this western cohort is something that merits further investigation. There are a number of potential explanations for this finding. Many of these patients may have subacute or chronic BCS, and outflow obstruction may not be complete. It is also possible that they had outpatient treatment with diuretics, also clinically insignificant ascites is probably not captured by the NIS database.

Hepatocellular carcinoma (HCC) is an important complication that is seen in long standing cases of BCS with incidence ranging from 2% to 40% depending on the geographical region.17, 18, 19, 20 According to one study, male sex, factor V Leiden mutation and IVC thrombosis were independent risk factors for developing HCC in BCS patients.¹⁸ In our study, 6.9% of hospitalized patients with BCS had HCC, and moreover, there was a significant increase in the prevalence of HCC from 1998 to 2017 (APC: 5.40%, P < 0.0001). On further analysis, we found that 43% of these patients had underlying viral hepatitis, and another 14% had alcoholic liver disease. The increasing prevalence of HCC during the 19-year period is concerning and corroborates with other studies that utilized the NIS.^21,22 Based on our observations, patients with BCS should have surveillance for HCC and it should include AFP which appears to have higher sensitivity to detect HCC in this subgroup of patients than other chronic liver diseases.¹⁸

Using a large administrative database such as the NIS comes with many limitations that must be acknowledged. First, we used ICD 9 and ICD 10 codes to capture the diagnosis of BCS, and this could result in coding errors potentially resulting in misclassification. We could not perform a sensitivity analysis because of the absence of patient identifiers in the data sets. Another major shortcoming is that the NIS reports every hospitalization as a separate encounter and not as a unique patient. It is possible many of these patients are readmitted and get counted multiple times leading to an overestimation of the actual prevalence. Using NIS data also may underestimate the disease prevalence because patients chronic or asymptomatic BCS are unlikely to be hospitalized. The database had undergone numerous changes over the 20-year period especially in the strategies used for sampling and weighting, including a significant design change in 2012. To account for these changes, for the years before 2011 we used revised discharge weights or TRENDWTs instead of the DISCWT to create national estimates for trend analysis.^23,24 Nonetheless, the NIS database is considered to be a powerful research tool providing robust clinical data about real world scenarios, and its reliability has been extensively validated. Despite some of these limitations, this is the largest population-based study on BCS from the USA. Our study shows that there is an increase in hospitalizations of patients with BCS, which may be reflective of an overall increase in disease prevalence in the country. Our observations also confirm the protean clinical manifestations of this rare syndrome using a very large national data set.

CRediT authorship contribution statement

Joseph J. Alukal: Conceptualization, Methodology, Funding acquisition, Formal analysis, Interpretation of the data, Writing - original draft, Writing - review & editing, Approved the final version, Agree to be accountable for all aspects of the work. Talan Zhang: Conceptualization, Methodology, Funding acquisition, Formal analysis, Interpretation of the data, Writing - original draft, Writing - review & editing, Approved the final version, Agree to be accountable for all aspects of the work. Paul J. Thuluvath: statistical analysis, Approved the final version, Agree to be accountable for all aspects of the work.

Conflicts of interest

The authors have none to declare.

Funding

None.

Appendix A. Supplementary data

The following is the Supplementary data to this article: