aDetroiHospital Cburgh MedStaten IslaArkansas,Medicine,manuscript

Drs. Pareport.

See pa*CorreE-mail

0002-9149http://dx.do

Results of Ventricular Septal Myectomy and HypertrophicCardiomyopathy (from Nationwide Inpatient Sample

[1998e2010])

Sidakpal S. Panaich, MDa, Apurva O. Badheka, MDa,*, Ankit Chothani, MDb, Kathan Mehta, MDc,Nileshkumar J. Patel, MDd, Abhishek Deshmukh, MDe, Vikas Singh, MDf,

Ghanshyambhai T. Savani, MDf, Shilpkumar Arora, MDa, Nilay Patel, MDa, Vipulkumar Bhalara, MDa,Peeyush Grover, MDf, Neeraj Shah, MDd, Mahir Elder, MDa, Tamam Mohamad, MDa, Amir Kaki, MDa,

Ashok Kondur, MDa, Michael Brown, MDa, Cindy Grines, MDa, and Theodore Schreiber, MDa

Ventricular septal myomectomy (VSM) is the primary modality for left ventricular outflow

t Medienter, Wdical Cnd UniLittle RMiami,receivenaich, B

ge 6 forspondinaddres

/14/$ -i.org/1

tract gradient reduction in patients with obstructive hypertrophic cardiomyopathy withrefractory symptoms. Comprehensive postprocedural data for VSM from a large multi-center registry are sparse. The primary objective of this study was to evaluate post-procedural mortality, complications, length of stay (LOS), and cost of hospitalization afterVSM and to further appraise the multivariate predictors of these outcomes. The HealthcareCost and Utilization Project’s Nationwide Inpatient Sample was queried from 1998 through2010 using International Classification of Diseases, Ninth Revision, procedure codes 37.33for VSM and 425.1 for hypertrophic cardiomyopathy. The severity of co-morbidities wasdefined using the Charlson co-morbidity index. Hierarchical mixed-effects models weregenerated to identify independent multivariate predictors of in-hospital mortality, proce-dural complications, LOS, and cost of hospitalization. The overall mortality was 5.9%.Almost 9% (8.7%) of patients had postprocedural complete heart block requiring pace-makers. Increasing Charlson co-morbidity index was associated with a higher rate ofcomplications and mortality (odds ratio 2.41, 95% confidence interval 1.17 to 4.98,p [ 0.02). The mean cost of hospitalization was $41,715 – $1,611, while the average LOSwas 8.89 – 0.35 days. Occurrence of any postoperative complication was associated withincreased cost of hospitalization (D$33,870, p <0.001) and LOS (D6.08 days, p <0.001). Inconclusion, the postoperative mortality rate for VSM was 5.9%; cardiac complications weremost common, specifically complete heart block. Age and increasing severity of co-morbidities were predictive of poorer outcomes, while a higher burden of postoperativecomplications was associated with a higher cost of hospitalization and LOS. � 2014Elsevier Inc. All rights reserved. (Am J Cardiol 2014;-:-e-)

Obstructive hypertrophic cardiomyopathy (HC) is acommon genetic disease with variable expressivity, char-acterized by varying degrees of left ventricular outflow tractobstruction.1 Ventricular septal myectomy (VSM), with>50 years of experience,2 has been the mainstay forameliorating outflow tract gradients in patients with medi-cally refractory symptoms.3,4 Published reports have

cal Center, Detroit, Michigan; bMedStar Washingtonashington, District of Columbia; cUniversity of Pitts-enter Shadyside Hospital, Pittsburgh, Pennsylvania;versity Hospital, Staten Island, New York; eUniversity ofock, Arkansas; and fUniversity of Miami Miller School ofFlorida. Manuscript received May 15, 2014; revisedd and accepted July 18, 2014.adheka, Chothani, and Mehta contributed equally to this

disclosure information.g author: Tel: (408) 324-4516; fax: (203) 737-2437.s: apurva_badheka@yahoo.com (A.O. Badheka).

see front matter � 2014 Elsevier Inc. All rights reserved.0.1016/j.amjcard.2014.07.075

demonstrated the efficacy of VSM in terms of improvementin hemodynamics and functional status as well as reductionsin syncope5 and sudden cardiac death after surgery.6e8

Furthermore, VSM offers the added advantage of correct-ing mitral valve apparatus abnormalities associated withHC.9 However, VSM is a complex procedure with a steeplearning curve that is best performed in the hands of expe-rienced operators at advanced tertiary centers with highvolumes.4,10 Most of the available data are limited toexperienced surgical centers with skilled high-volume op-erators.7,11 Real-world data for postprocedural outcomes forVSM from multiple centers across the nation are sparse.Concern regarding postprocedural outcomes has becomeincreasingly germane given an increase in the number ofreferred patients as well the availability of a less invasivealternative (alcohol septal ablation).10 The main objectivesof our study were (1) to evaluate the postsurgical outcomesof VSM in terms of mortality as well as complications, (2)to further elucidate potential predictors of post-operativeoutcomes, and (3) to study resource utilization in terms of

www.ajconline.org

Table 1Baseline characteristics of patient undergoing ventricular septalmyomectomy in United States from 1998e2010 (n ¼ 665)

Age (year) (Mean � SE) 56.9 � 0.6Male 40.0%White 52.8%Non-white 11.7Charlson/deyo comorbidity index* (Mean � SE) 0.87 � 0.04Obesity (Body Mass Index �30) 11%Hypertension 45.1%Diabetes mellitus 12.2%Heart failure 0.6%Chronic pulmonary disease 15.8%Peripheral vascular disease 4.4%Renal failure 3.2%Neurological disorder/paralysis 2.1%Anemia/coagulopathy 22.3%Hematological or oncological malignancy 0.8%Weight loss/cachexia 1.8%Collagen vascular disease 1.5%Depression/substance abuse 9.2%Median household income category for patient’s zip code†

0e25th percentile 20.5%26e50th percentile 23%51e75th percentile 19.1%76e100th percentile 15.5%

Primary PayerMedicare/Medicaid 46.3%Private including HMO 48.6%Self pay/no charge/other 5%Missing 0.2%

Hospital bed sizeSmall 5.7%Medium 11.4%Large 82.3%Missing 0.6%

Hospital LocationUrban 97.3%Missing 0.6%

Hospital RegionNortheast 16.5%Mid West/North Central 34.3%South 30.4%West 18.8%Missing

Hospital Teaching statusTeaching 77.7%Missing 0.6%

Admission typesEmergent/Urgent 19.7%Missing 8.3%

Admission dayWeekend 4.7%

Length of stay (Days) (Means � SE) 8.89 � 0.35Cost ($) (Means � SE) 41,715 � 1,611Missing 1.3%

Peri-procedural complications 30.0%DispositionHome 79.7%Facility 14.4%Missing 5.9%

Death 5.9%

Race was missing in 35.5% of population.HMO ¼ health maintenance organization; SE ¼ standard error.

2 The American Journal of Cardiology (www.ajconline.org)

length of stay (LOS) and cost of hospitalization associatedwith this surgical procedure by analyzing the largest pub-licly available inpatient care database.

Methods

The study cohort was derived from the NationwideInpatient Sample (NIS) database from 1998 through 2010, asubset of the Healthcare Cost and Utilization Project spon-sored by the Agency for Healthcare Research and Quality.The NIS is the largest publicly available all-payer inpatientcare database in the United States, including data onapproximately 7 million to 8 million discharges per year,and is a stratified sample designed to approximate a 20%sample of US community (nonfederal, short-term, general,and specialty) hospitals. National estimates are producedusing sampling weights provided by the sponsor. Detailsregarding the NIS data have been previously published.Overall, the NIS contains about 8 million inpatients annu-ally. Annual data quality assessments of the NIS are per-formed, which guarantee the internal validity of thedatabase. Furthermore, comparisons against the followingdata sources strengthen the external validity of the NIS: theAmerican Hospital Association Annual Survey Database,the National Hospital Discharge Survey from the NationalCenter for Health Statistics, and Medicare Provider Analysisand Review inpatient data from the Centers for Medicareand Medicaid Services.

We queried the NIS database from 1998 through 2010using International Classification of Diseases, Ninth Revi-sion, Clinical Modification (ICD-9-CM) procedure codes37.33 for VSM and 425.1 for HC. NIS variables were used toidentify patients’ demographic characteristics, including age,gender, and race (Table 1). We defined the severity of co-morbid conditions using the Deyo modification of theCharlson co-morbidity index (CCI) (Supplementary Table 1).This index contains 17 co-morbid conditions with differentialweights. The score ranges from 0 to 33, with higher scorescorresponding to a greater burden of co-morbid diseases.Preventable procedural complications were identified by pa-tient safety indicators (PSIs), which have been established bythe AHRQ to monitor preventable adverse events duringhospitalization. These indicators are based on ICD-9-CMcodes and Medicare severity diagnosis-related groups, andeach PSI has specific inclusion and exclusion criteria. Indi-vidual PSI technical specifications were used to identifyand define preventable procedural complications, namely,

* Charlson/Deyo comorbidity index was calculated as per - Deyo RA,Cherkin DC, Coil MA. Adapting a clinical comorbidity index for use withICD-9-CM administrative databases. J ClinEpidemiol.1992;45:613e619.

† This represents a quartile classification of the estimated medianhousehold income of residents in the patient’s ZIP Code. These values arederived from ZIP Code-demographic data obtained from Claritas. Thequartiles are identified by values of 1 to 4, indicating the poorest towealthiest populations. Because these estimates are updated annually, thevalue ranges vary by year.zip is missing in 21.9% of population. http://www.hcup- us.ahrq.gov/db/vars/zipinc_qrtl/nisnote.jsp.

Table 2Post-procedural complication related to ventricular septal myomectomy in United States from 1998e2010

Variable ICD CODE Percentage

Death 5.9%Any procedural complications 30.2%Death þ Any procedural complications 31.4%Vascular complications 9.6%Postop-hemorrhage requiring transfusion* 998.11, 998.12, 99.0, V58.2 5.4%Vascular complications including - Injury to blood vessels-900-904

- Accidental puncture-998.2, e8700-8709 (PSI)- AV fistula-447- Injury to retro-peritoneum 8680.4- Vascular complications requiring surgery-39.31, 39.41, 39.49, 39.52,39.53, 39.56, 39.57, 39.58, 39.59, 39.79

- Other vascular complications-999.2, 997.7

4.4%

Cardiac complications 15.9%Iatrogenic cardiac complications 997.1 10.5%CHB requiring pacemaker Insertion 37.80e83 8.7%Pericardial complications 423.0-Hemopericardium 0.2%

423.3-Cardiac temponad337.0-Pericardiocentesis

Respiratory complications (Post-op respiratory failure) 512.1 3.9%Neurological ComplicationsPostop-Stroke/TIA 997.0, 997.00, 997.01, 997.02, 435.9,438.0, 4381.0, 4381.1, 4381.2,

4381.9, 4382.0, 4382.1, 4382.2, 4383.0, 4383.1, 4383.2, 4384.0,4384.1, 4384.2, 4385.0, 4385.1, 4385.2, 4385.3, 4388.1, 4388.2,4388.9, 438.9

2.6%

Renal and metabolic complications 1.4%Postoperative DVT/PE PSI† 1.5%Postop infectious complicationsz PSI† 3.3%Pressure ulcer rate PSI† 0.6%

* Hemorrhage requiring transfusion was identified as having any patient having post-operative hemorrhage and also got transfusion.† Post-procedural complications were identified by Patient Safety Indicators (PSIs) which have been established by the Agency for Healthcare Research and

Quality to monitor preventable adverse events during hospitalization. These indicators are based on ICD-9-CM codes and Medicare severity Diagnosis-RelatedGroups and each PSI has specific inclusion and exclusion criteria. PSI individual measure technical specifications, Version 4.4, March 2012 was used toidentify & define preventable complications.

z Infectious complication were identified as composite of post-operative sepsis, septic shock or catheter related infection.

Cardiomyopathy/Ventricular Septal Myectomy and Hypertrophic Cardiomyopathy 3

postprocedural respiratory failure, postprocedural renal andmetabolic derangement, postprocedural pulmonary embolismor deep vein thrombosis, procedural infectious complications(including postprocedure sepsis and central venous catheter-erelated bloodstream infection), pressure ulcers, and acci-dental puncture or laceration. Other procedure-relatedcomplications, including hemorrhage requiring blood trans-fusion, iatrogenic cardiac complications, implantation of apermanent pacemaker implying high-grade heart block,pericardial complications, conversion to open-heart surgery,other iatrogenic cardiac complications (including coronarydissections and chamber perforations), procedural stroke ortransient ischemic attack, and vascular complications, wereidentified using ICD-9-CM codes in any secondary diagnosisfield (Supplementary Table 2). To prevent classification of apreexisting condition (e.g., stroke or heart block) as acomplication, cases with the ICD-9-CM code for a compli-cation listed as the principal diagnosis were excluded.Vascular complications were defined as the PSI code foraccidental puncture or the ICD-9-CM codes for injury toblood vessels, creation of an arteriovenous fistula, injury toretroperitoneum, vascular complications requiring surgery,

and other vascular complications not elsewhere classified.“Any complications” was defined as the occurrence of �1procedural complications. This method has been used inearlier studies.12

Annual hospital volume was determined on a year-to-year basis using the unique hospital identification numberto calculate the total number of procedures performed by aparticular institution in a given year. Similarly, operatorvolume was computed using operator identification number,although not all hospitals report operator identificationnumbers. Furthermore, operator identification numbers werenot reported in the NIS for 2010. Hospital volume wasincorporated as a continuous variable in the multivariatemodel in increments of 3 units per year. Operator volumewas similarly incorporated in 1-unit increments. Multivar-iate models were created incorporating hospital and operatorvolume, with a term to adjust for the interaction effect be-tween hospital and operator volume. Hospital identity wasincorporated as a random effect in the model to account forthe effect of hospital clustering (meaning that patientstreated at the same hospital may experience similar out-comes as a result of other processes of care).

Table 3Multivariable predictors of mortality or complications in the studypopulation for hypertrophic cardiomyopathy ablation (N ¼ 194)

Variable Model

OR (95% CI) p Value

Age 1.04 (1.01e1.07) 0.01Female Gender 0.60 (0.32e1.14) 0.1Deyo Modification of CCI*¼0 Referent�1 2.41 (1.17e4.98) 0.02

Primary PayerMedicare/Medicaid ReferentPrivate Insurance 1.06 (0.46e2.44) 0.9Other/No insurance 0.34 (0.04e2.61) 0.3

Teaching Hospital Status 0.44 (0.16e1.23) 0.1Operator volume 0.78 (0.39e1.54) 0.5Hospital volume 0.80 (0.27e2.35) 0.7

* CCI ¼ Charlson co-morbidity index.

Table 4Multivariable predictors of cost of hospitalization and length of stay forhypertrophic cardiomyopathy ablation

Variable Model for Length ofStay (N ¼ 291)

Model for Cost(N ¼ 181)

Estimate ofLength of

Hospitalization(Days)

pValue

Estimate ofCost (Dollars)

pValue

Intercept þ5.29 0.2 þ28,856 0.2Age þ0.08 0.1 þ274 0.2Female Gender þ0.45 0.7 �3,891 0.4Deyo Modification

of CCI*¼0 Referent Referent�1 �1.53 0.1 �4,081 0.4

Primary PayerMedicare/Medicaid Referent ReferentPrivate Insurance �1.51 0.2 �8,961 0.1Other �1.04 0.7 þ8,092 0.6

Teaching Hospital Status �1.58 0.2 þ6,590 0.3Operator Volume �0.24 0.8 �2,644 0.5Hospital Volume

3 unit increase�0.20 0.9 �1,070 0.9

Any Complications þ6.08 <0.001 þ33,870 <0.001

* CCI ¼ Charlson co-morbidity index.

4 The American Journal of Cardiology (www.ajconline.org)

The total duration of hospital stay in days was estimatedfor all patients, after excluding those who died in the hos-pital, using the information on LOS provided in the NIS dataset. The NIS data set includes all patients admitted underobservational or inpatient status into participating hospitals.The NIS contains data on total charges for each hospital inthe databases, which represents the amounts hospitals billedfor services. To calculate estimated cost of hospitalizations,the NIS data were merged with cost-to-charge ratios avail-able from the Healthcare Cost and Utilization Project. Usingthe merged data elements from the cost-to-charge ratio filesand the total charges reported in the NIS database, weconverted the hospital total charge data to cost estimates by

simply multiplying total charges by the appropriate cost-to-charge ratio. These costs are in essence standardized and canbe measured across hospitals and are used in the remainderof this report. Adjusted cost for each year was calculated interms of the 2010 cost, after adjusting for inflation accordingto the latest Consumer Price Index data released by the USgovernment on January 16, 2013.

Stata IC version 11.0 (StataCorp LP, College Station,Texas) and SAS version 9.0 (SAS Institute Inc., Cary, NorthCarolina) were used for analyses. Weighted values ofpatient-level observations were generated to produce a na-tionally representative estimate of the entire US populationof hospitalized patients. Differences between categoricalvariables were tested using chi-square tests, and differencesbetween continuous variables were tested using Student’st tests. A p value <0.05 was considered significant. The NISdata set is inherently hierarchical; that is, the data havegroup (i.e., hospital)especific attributes, and within eachgroup (i.e., hospital) there are patients who contributepatient-specific attributes to the data. Hierarchical modelstake into consideration the effect of nesting (e.g., patient-level effects nested within hospital-level effects). Hence,hierarchical modeling is superior to simple regressionmodeling for the available data set. Two-level hierarchicalmodels (with patient-level factors nested within hospital-level factors) were created with the unique hospital identi-fication number incorporated as a random effect within themodel. Hierarchical mixed-effects logistic regressionmodels were used for categorical dependent variables suchas in-hospital mortality and procedural complications, andhierarchical mixed-effects linear regression models wereused for continuous dependent variables such as cost ofhospitalization and LOS. Variables with >10% missing datawere not included in the multivariate models. In all multi-variate models, we included hospital-level variables, such ashospital region (Northeast, South, Midwest, and West [thereferent]) and teaching versus nonteaching status, andpatient-level variables, such as age, gender, Deyo modifi-cation of CCI, occurrence of procedural complications,admission over the weekend, and primary payer (withMedicare or Medicaid as the referent), in addition to hospitaland operator procedure volume. All interactions were thor-oughly tested. Multicollinearity, defined as a perfect linearrelation or a very high correlation between �2 predictor(independent) variables, was assessed using varianceinflation factors, with values >20 suggestive ofmulticollinearity.

Results

Table 1 shows baseline characteristics of the studypopulation. A total of 665 VSM procedures were availablefor analysis from 1998 to 2010. The mean age of the studycohort was 56.9 � 0.6 years. Men constituted 40% of thecohort, with 52.8% being white. The mean CCI score forthe cohort was calculated as 0.87 � 0.04, with hypertensionbeing the most common co-morbidity, present in 45.1% ofthe patients, while diabetes was present in 12.2% of thosewho underwent VSM. Most procedures were done at large(82.3%), urban (97.3%), or teaching (77.7%) hospitals.Almost 80% (79.7%) of patients were discharged home

Cardiomyopathy/Ventricular Septal Myectomy and Hypertrophic Cardiomyopathy 5

after the procedure, while 14.4% were discharged tofacilities.

The overall postprocedural mortality (Table 2) was 5.9%,while the rate of postprocedural complications was 30.2%(Table 2). Cardiac complications were most common(15.9%), including iatrogenic cardiac complications (10.5%)and complete heart block requiring pacemaker insertion(8.7%). Vascular complications, including access-site com-plications, occurred in 9.6% of patients, of whom 5.4%required transfusion. Respiratory complications occurred in3.9% of patients, while 1.4% of patients had renal ormetabolic complications.

Increasing burden of co-morbidities (Table 3) as repre-sented by higher CCI score was associated with a higher rateof complications and mortality (odds ratio [OR] 2.41, 95%confidence interval [CI] 1.17 to 4.98, p ¼ 0.02). Age was asignificant predictor of postprocedural mortality and com-plications (OR 1.04, 95% CI 1.01 to 10.7, p ¼ 0.01). Higheroperator (OR 0.78, 95% CI 0.39 to 1.54, p ¼ 0.50) andhospital (OR 0.80, 95% CI 0.27 to 2.35, p ¼ 0.70) volumeswere associated with a lower rate of complications andmortality, although neither reached statistical significance.There has been a significant decrease in mortality (p valuefor temporal trend ¼ 0.04) in recent years, plausibly due toimproved surgical techniques (Supplementary Figure 1).

The average LOS was 8.89 � 0.35 days, while the meancost of hospitalization was $41,715 � $1,611. The onlyindependent predictors of both LOS and cost of hospitali-zation were any procedural complications (Table 4). In theevent of any complications, the average LOS was increasedby 6.08 days (p <0.001), while the average cost of hospi-talization was increased by $33,870 (p <0.001).

Discussion

Our study presents valuable real-world postproceduraldata regarding VSM from the largest available inpatient caredatabase. Postoperative mortality after VSM was noted to behigh at 5.9% in our study on the basis a wide spectrum ofreal-world VSM experience from multiple communityhospitals across the nation. Although earlier publicationshave reported similar13 and higher14 mortality rates,improving surgical experience has resulted in reductions inpostoperative mortality and complication rates in recentyears. Our results are thus in contradistinction to recentlyreported single-center experiences from high-volume hos-pitals with low postoperative mortality of <1%.4,15 Thisreiterates an important caveat that replication of superiorsurgical results from higher volume centers with institu-tional commitment to VSM might not be possible at otherlower volume facilities or hospitals.16 Although there was atrend toward lower postoperative mortality and complica-tions with improving operator and hospital volumes, wecould not demonstrate statistical significance. This couldplausibly be secondary to a lack of sufficient events ormissing operator identification numbers by certain hospitals,confounding the final results. Future registries with longerfollow-up periods and/or larger sample sizes might beconstructive in sorting out these differences.

Among postoperative cardiac complications, 8.7% ofpatients required pacemaker insertion for complete heart

block, a rate similar to that reported by Smedira et al17

(7.9%) in their registry from the Cleveland Clinic. Higherrates of pacemaker implantation have been reported inearlier studies,18 although recent studies have noted re-ductions in pacemaker implantation rates at specializedcenters.10 Although patients without preexisting conductionblocks are known to have lower rates of pacemaker im-plantation after VSM,17 details regarding preproceduralconduction disturbances were not available in our cohort.Pericardial complications (0.2%) were also rare and similarto previous reports (0.93%). Postoperative neurologic(2.6%) and infectious complications (3.3%) were slightlyhigher than previously reported results (0.62%).17 Our studyfurther included a much more exhaustive list of complica-tions compared with previous studies, including respiratory,vascular, renal, and metabolic complications.

Age was a significant predictor of mortality and post-operative complications in our study. This parallels thepublished research regarding higher mortality in older pa-tients14,19,20 after VSM in addition to patients with co-morbid conditions.6,11,21 Indeed, patients with a higherburden of co-morbidities had higher mortality and worsepostprocedural outcomes in our study. VSM yields the bestclinical outcomes and mortality benefit in younger patientswith fewer co-morbidities.22 Advanced age is also known toindependently predict procedural and delayed conductionblocks.23 Our results further reiterate the importance ofclosely scrutinized selection of patients who are at a highersurgical risk because of age and especially baseline co-morbidities. Although our study did not achieve statisti-cally significant results in outcomes on the basis of operatorvolumes, the comparison of our results with previous single-center experiences indicates that this could perhaps be bestachieved at higher volume centers by physicians with spe-cial interest in treating primarily patients with HC.

Ours is one of the very few studies reporting resourceutilization for VSM. The average LOS of 8.89 � 0.35 dayswas relatively similar to that reported for Toronto GeneralHospital (median duration 7 days, range 3 to 19).24 The onlysignificant predictor of LOS and hospitalization costs wasthe occurrence of postoperative complications. This is un-derstandable given the lengthy recovery of patients in caseof complications, which would drive up hospitalizationcosts as well. The current burden of health care costs onnational finances mandates scrutiny of procedural outcomesand postprocedural care to avoid any potential complica-tions that could result in escalation of hospitalization ex-penditures. This again forces one to consider whetherrestricting complex surgical procedures to higher volumecenters with dedicated care for such patients would result inbetter outcomes with improved resource utilization. Furtherstudies comparing outcomes between dedicated HC referralcenters and lower volume hospitals might be of value.

An important limitation of our study is related to the posthoc analysis of large administrative data sets, which can beassociated with coding errors including underreporting ofsecondary or co-morbid diagnosis in addition to a lack oflong-term follow-up data, as well as information on post-procedural hemodynamic and echocardiographic variables.Furthermore, the procedure and operator volumes could bemisrepresented by the sample, especially if the operators

6 The American Journal of Cardiology (www.ajconline.org)

perform VSM at hospitals that are not part of the NISdatabase. Thus, there is a trade-off between using smallercohorts with more refined clinical information and admin-istrative databases with larger sample sizes, although thelatter is an important source of population-based informa-tion on procedural outcomes. Finally, we lack data onconcomitant cardiac surgical procedures such as mitral valverepair that might have been performed along with VSM,thus plausibly affecting postprocedural mortality andcomplications.

Disclosures

The authors have no conflicts of interest to disclose.

Supplementary Data

Supplementary data associated with this article can befound, in the online version, at http://dx.doi.org/10.1016/j.amjcard.2014.07.075.

1. Maron MS, Olivotto I, Zenovich AG, Link MS, Pandian NG, Kuvin JT,Nistri S, Cecchi F, Udelson JE, Maron BJ. Hypertrophic cardiomy-opathy is predominantly a disease of left ventricular outflow tractobstruction. Circulation 2006;114:2232e2239.

2. Cleland WP. The surgical management of obstructive cardiomyopathy.J Cardiovasc Surg (Torino) 1963;4:489e491.

3. Maron BJ, McKenna WJ, Danielson GK, Kappenberger LJ, Kuhn HJ,Seidman CE, Shah PM, Spencer WH 3rd, Spirito P, Ten Cate FJ, WigleED. American College of Cardiology/European Society of Cardiologyclinical expert consensus document on hypertrophic cardiomyopathy.A report of the American College of Cardiology Foundation Task Forceon Clinical Expert Consensus Documents and the European Society ofCardiology Committee for Practice Guidelines. J Am Coll Cardiol2003;42:1687e1713.

4. Williams LK, Rakowski H. Surgical myectomy for hypertrophicobstructive cardiomyopathy: the cut that heals. Circulation 2013;128:193e197.

5. Orme NM, Sorajja P, Dearani JA, Schaff HV, Gersh BJ, Ommen SR.Comparison of surgical septal myectomy to medical therapy alone inpatients with hypertrophic cardiomyopathy and syncope. Am J Cardiol2013;111:388e392.

6. Desai MY, Bhonsale A, Smedira NG, Naji P, Thamilarasan M, LytleBW, Lever HM. Predictors of long-term outcomes in symptomatichypertrophic obstructive cardiomyopathy patients undergoing surgicalrelief of left ventricular outflow tract obstruction. Circulation2013;128:209e216.

7. Ommen SR, Maron BJ, Olivotto I, Maron MS, Cecchi F, Betocchi S,Gersh BJ, Ackerman MJ, McCully RB, Dearani JA, Schaff HV,Danielson GK, Tajik AJ, Nishimura RA. Long-term effects of surgicalseptal myectomy on survival in patients with obstructive hypertrophiccardiomyopathy. J Am Coll Cardiol 2005;46:470e476.

8. McLeod CJ, Ommen SR, Ackerman MJ, Weivoda PL, Shen WK,Dearani JA, Schaff HV, Tajik AJ, Gersh BJ. Surgical septal myectomydecreases the risk for appropriate implantable cardioverter defibrillatordischarge in obstructive hypertrophic cardiomyopathy. Eur Heart J2007;28:2583e2588.

9. Minakata K, Dearani JA, Nishimura RA, Maron BJ, Danielson GK.Extended septal myectomy for hypertrophic obstructive cardiomyopa-thy with anomalous mitral papillary muscles or chordae. J ThoracCardiovasc Surg 2004;127:481e489.

10. Iacovoni A, Spirito P, Simon C, Iascone M, Di Dedda G, De Filippo P,Pentiricci S, Boni L, Senni M, Gavazzi A, Ferrazzi P. A contemporaryEuropean experience with surgical septal myectomy in hypertrophiccardiomyopathy. Eur Heart J 2012;33:2080e2087.

11. Woo A, Williams WG, Choi R, Wigle ED, Rozenblyum E, Fedwick K,Siu S, Ralph-Edwards A, Rakowski H. Clinical and echocardiographicdeterminants of long-term survival after surgical myectomy inobstructive hypertrophic cardiomyopathy. Circulation 2005;111:2033e2041.

12. Deshmukh A, Patel NJ, Pant S, Shah N, Chothani A, Mehta K, GroverP, Singh V, Vallurupalli S, Savani GT, Badheka A, Tuliani T, Dab-hadkar K, Dibu G, Reddy YM, Sewani A, Kowalski M, Mitrani R,Paydak H, Viles-Gonzalez JF. In-hospital complications associatedwith catheter ablation of atrial fibrillation in the United States between2000 and 2010: analysis of 93 801 procedures. Circulation 2013;128:2104e2112.

13. Firoozi S, Elliott PM, Sharma S, Murday A, Brecker SJ, Hamid MS,Sachdev B, Thaman R, McKenna WJ. Septal myotomy-myectomy andtranscoronary septal alcohol ablation in hypertrophic obstructive car-diomyopathy. A comparison of clinical, haemodynamic and exerciseoutcomes. Eur Heart J 2002;23:1617e1624.

14. Heric B, Lytle BW, Miller DP, Rosenkranz ER, Lever HM, CosgroveDM. Surgical management of hypertrophic obstructive cardiomyopa-thy. Early and late results. J Thorac Cardiovasc Surg 1995;110:195e206; discussion 206e208.

15. Maron BJ, Dearani JA, Ommen SR, Maron MS, Schaff HV, Gersh BJ,Nishimura RA. The case for surgery in obstructive hypertrophic car-diomyopathy. J Am Coll Cardiol 2004;44:2044e2053.

16. Elmariah S, Fifer MA. Medical, surgical and interventional manage-ment of hypertrophic cardiomyopathy with obstruction. Curr TreatOptions Cardiovasc Med 2012;14:665e678.

17. Smedira NG, Lytle BW, Lever HM, Rajeswaran J, Krishnaswamy G,Kaple RK, Dolney DO, Blackstone EH. Current effectiveness and risksof isolated septal myectomy for hypertrophic obstructive cardiomy-opathy. Ann Thorac Surg 2008;85:127e133.

18. Qin JX, Shiota T, Lever HM, Asher CR, Popovic ZB, Greenberg NL,Agler DA, Drinko JK, Smedira NG, Tuzcu EM, Lytle BW, ThomasJD. Conduction system abnormalities in patients with obstructive hy-pertrophic cardiomyopathy following septal reduction interventions.Am J Cardiol 2004;93:171e175.

19. Cooper MM, McIntosh CL, Tucker E, Clark RE. Operation for hy-pertrophic subaortic stenosis in the aged. Ann Thorac Surg 1987;44:370e378.

20. Schulte HD, Bircks WH, Loesse B, Godehardt EA, Schwartzkopff B.Prognosis of patients with hypertrophic obstructive cardiomyopathyafter transaortic myectomy. Late results up to twenty-five years.J Thorac Cardiovasc Surg 1993;106:709e717.

21. Ross RE, Sherrid MV, Casey MM, Swistel DG, Balaram SK. Doessurgical relief of obstruction improve prognosis for hypertrophic car-diomyopathy? Prog Cardiovasc Dis 2012;54:529e534.

22. Sorajja P, Valeti U, Nishimura RA, Ommen SR, Rihal CS, Gersh BJ,Hodge DO, Schaff HV, Holmes DR Jr. Outcome of alcohol septalablation for obstructive hypertrophic cardiomyopathy. Circulation2008;118:131e139.

23. Lawrenz T, Lieder F, Bartelsmeier M, Leuner C, Borchert B, Meyer zuVilsendorf D, Strunk-Mueller C, Reinhardt J, Feuchtl A, Stellbrink C,Kuhn H. Predictors of complete heart block after transcoronary ablationof septal hypertrophy: results of a prospective electrophysiologicalinvestigation in 172 patients with hypertrophic obstructive cardiomy-opathy. J Am Coll Cardiol 2007;49:2356e2363.

24. Ralph-Edwards A, Woo A, McCrindle BW, Shapero JL, Schwartz L,Rakowski H, Wigle ED, Williams WG. Hypertrophic obstructive car-diomyopathy: comparison of outcomes after myectomy or alcoholablation adjusted by propensity score. J Thorac Cardiovasc Surg2005;129:351e358.