Impact of Complications on Outcomes Following Aortic and Mitral Valve Replacements in the United States

Written By: Sam Savage

Published Date: September 25, 2007 Last Edited: September 25, 2007

By Allareddy, V Ward, M M; Ely, J W; Allareddy, V; Levett, J

Aim. Heart valve replacement surgeries account for 20% of all cardiac procedures. In-hospital mortality rates are approximately 6% for aortic valve replacements and 10% for mitral valve replacements. The objectives of the study are to provide nationally representative estimates of complications following aortic and mitral valve replacements and to quantify the impact of different types of complications on in-hospital outcomes. Methods. The Nationwide Inpatient Sample was analyzed for years 2000-2003. The effect of complications on in-hospital mortality, length of stay (LOS), and hospital charges were examined using bivariate and multivariable logistic and linear regression analyses. The confounding effects of age, sex, primary diagnosis, type of valve replacement, type of admission, comorbid conditions, and hospital characteristics were adjusted.

Results. A total of 43 909 patients underwent aortic valve replacement as the primary procedure during the study period and 16 516 patients underwent mitral valve replacement. Complications occurred in 35.2% of those undergoing aortic valve replacements and in 36.4% of those undergoing mitral valve replacements. Almost half of these are cardiac complications and a quarter involve hemorrhage/ hematoma/seroma. Complications were significantly associated with in- hospital mortality, LOS, and hospital charges even after adjusting for patient and hospital characteristics.

Conclusion. Complications are prevalent and exert a considerable influence on outcomes following aortic and mitral valve replacements. Quality initiatives should focus on minimizing complications and improving processes of care that would enable complications to be better resolved if they occur.

KEY WORDS: Aortic valve, surgery – Mitral valve, surgery – Prosthesis – Treatment, outcome.

Aortic and mitral valve replacements are often treatments of choice for patients with symptomatic aortic and mitral valve diseases respectively. Heart valve replacement surgeries account for 20% of all cardiac procedures and as the proportion of elderly in the United States increases, the number of valve replacements is expected to rise.1 Heart valve replacements also account for 30% of all deaths following cardiac surgeries l with in-hospital mortality rates around 4.3% for first-time isolated aortic valve replacements 2 and 10% for mitral valve replacements and surgeries. 3 In- hospital mortality rates are higher for multiple valve replacements when compared to isolated aortic valve replacements.2 Several single center and population-based studies have examined the predictors of mortality following valve replacement surgeries.2-10 Most studies reporting the incidence of complications following heart valve replacement surgeries are single center trials.11-14 Single center trials, though potentially rich in individual patient data, do not accurately represent usual practice because their results may not be representative of other hospitals. There is little published data providing population-based estimates of complications following heart valve replacement surgeries or describing the impact of complications on outcomes.

The objectives of the current study are to provide nationally representative estimates of complications following aortic and mitral valve replacement surgeries and to quantify the relative impact of different types of complications on in-hospital outcomes including in-hospital mortality, length of stay (LOS), and hospital charges.

Materials and methods

Patient selection

Analyses were conducted using the Nationwide Inpatient Sample (NIS) of the Healthcare Cost and Utilization Project (HCUP) for the years 2000-2003. The NIS is a 20% stratified sample of all non- federal hospitals in the United States and contains up to 8 million records from approximately 1 000 hospitals in 35 states.15 All patients who underwent aortic valve replacement (ICD-9-CM codes of ‘35.21’ or ‘35.22’) or mitral valve replacement (ICD-9-CM codes of ‘35.23’ or ‘35.24’) as primary procedures were selected for analyses.

Complications

The NIS dataset has information on up to 15 diagnoses per discharge. Reported complications were classified using the Clinical Classification Software (CCS) coding system (code = 238 – Complications due to medical or surgical care) in the NIS. Based on the ICD-9-CM codes, we grouped the complications into 12 categories including: cardiac, nervous, respiratory, digestive, urinary, iatrogenic, infections, hemorrhage/hematoma/seroma,disruption/ dehiscence/rupture/non-healing wounds, septicemia, and others. The ICD-9-CM codes for complications that were used in this study are summarized in Table I.

Outcome variables

The hospital charges in the NIS were adjusted to year 2003 (last quarter) levels using the consumer price index for inpatient hospital services provided by the Bureau of Labor Statistics.16 Hospital charges were log transformed to correct for skewed data distribution as was LOS. In-hospital mortality was coded in the dataset.

Descriptive statistical snalysis

Descriptive statistics were used to summarize the frequency of occurrence of complications and the characteristics of patients with and without complications. For the procedures of interest, baseline characteristics such as age, sex, admission type (elective versus emergent/urgent), type of valve replacement (bioprosthetic or mechanical valve) and performance of concomitant coronary artery bypass graft (CABG) procedure during the hospitalization were compared between those who developed complications and those who did not develop complications by using independent sample t tests and ?2 tests where appropriate. Hospital volume was used as a continuous variable in the models and was calculated based on the number of aortic and mitral valve replacement procedures performed (includes secondary procedures) at each hospital each year.

Regression models

The influence of patient related factors including age, sex, type of valve replacement procedure, multiple valve replacement (concomitant aortic or mitral valve), presence of comorbid conditions, admission type, primary diagnosis, performance of CABG, number of complications, and different types of complications were examined on in-hospital mortality, LOS, and hospital charges following aortic and mitral valve replacement surgeries using bivariate logistic and linear regression models where appropriate.

Multivariable logistic and linear regression models were then developed incorporating the variables that were significant (P

TABLE I.-Frequencies of complications following aortic valve or mitral valve replacement.

In the NIS, 11 states do not report race,1? which means that this variable is missing from 30% of cases. Thus, to retain the full N in the multivariable models, they were computed once without race in the model and then computed again, on a reduced sample size, with race in the model.

Statistical analysis

The Generalized Estimating Equation (GEE) method specifying an exchangeable correlation matrix was used to fit the bivariate and multivariable regression models in order to correct for possible clustering of similar outcomes within hospitals.18 The empirical standard errors were used to compute the 95% confidence intervals of the estimates. Two-sided P values were calculated for all analyses and 0.05 was set as the statistical significance level. All analyses were performed using SAS version 9.1 (SAS Institute Ine, Cary, NC) and SPSS version 13.0 (SPSS Ine, Chicago, IL).

Results

A total of 43 909 patients underwent aortic valve replacement as the primary procedure during the study period and 16 516 patients underwent mitral valve replacement. The baseline characteristics of patients undergoing these procedures are summarized in Table II. The average age of patients having aortic valve replacement was 67 years and 64 years for mitral valve replacement surgery. Nearly 40% of the patients who underwent aortic valve replacement and 56% of patients who underwent mitral valve replacement were female. Approximately 66% of admissions for aortic valve replacement and 6l% of admissions for mitral valve replacement were elective. Mechanical valves were used for 56% and 71% of patients having aortic valve and mitral valve replacements respectively. The most frequently occurring primary diagnosis for both aortic and mitral valve replacement surgeries was heart valve disorders. Patients who were transferred to another short-term hospital (456 after aortic valve and 197 after mitral valve replacement) or for whom information about disposition was not available (158 after aortic valve replacement and 52 after mitral valve replacement) were not included in the analysis. TABLE II.-Baseline characteristics of patients undergoing aortic valve and mitral valve replacement surgeries.

TABLE III.-Characteristics of patients* with and without complications after aortic valve or mitral valve replacement.

Close to 35% of those undergoing aortic valve replacements and 36.4% of those undergoing mitral valve replacements had at least one complication. The comparative baseline characteristics of patients who developed and those who did not develop complications following heart valve replacement surgeries are summarized in Table III. Overall, complications were seen more frequently among older patients (P

The frequencies of different types of complications are shown in Table I. Cardiac complications were the most common, affecting 16.9% of patients with aortic valve replacement and 15.5% of patients with mitral valve replacement. Hemorrhage /hematoma /seroma was also common, occurring in 8.0% of patients with aortic valve replacement and 8.3% of patients with mitral valve replacement, with respiratory complications next most common (affecting 5.0% and 5.6% respectively). Other complications, including iatrogenic induced complications, urinary complications, nervous system complications, infections, and septicemia occurred in 1 to 5% of patients. Relatively rare complications (occurring in

In-hospital mortality, median LOS, and median hospital charges for the various types of complications are shown in Table IV. In- hospital mortality rates among those who developed complications following aortic valve replacement were 10% and 13% following mitral valve replacement. The median LOS for patients with complications was 10 days for aortic valve replacement and 12 days for mitral valve replacement. The median hospital charge for patients who developed complications following aortic valve replacement was $24,000 more than those who did not develop a complication. The corresponding number for those who underwent mitral valve replacement was $30,690.

The results of the multivariable analyses for in-hospital mortality are shown in Table V. In the bivariate analyses, the odds of in-hospital mortality for both aortic and mitral valve replacements increased with an increase in number of complications. For aortic valve replacement, all complications except vascular complications were associated with higher in-hospital mortality rates (P

In the multivariable models, after adjusting for all patient characteristics, patients who developed two or more complications following aortic valve replacement had 305% higher odds (P

TABLE IV.-Complications and their associated mortality rates, median LOS, and median hospital charges following aortic valve or mitral valve replacement.

The results of analyses for LOS and hospital charges are generally similar to those of mortality. In the Divariate analyses for both aortic and mitral valve replacements, number of complications, type of complications, and all patient characteristics were significantly associated with increased LOS. In the multivariable analyses, LOS increased with an increase in the number of complications for both aortic and mitral valve replacements (P

In the bivariate analyses, hospital charges increased with an increase in the number of complications and all complication types were associated with higher hospital charges for both aortic and mitral valve replacements (P

Complications continued to be significantly associated with in- hospital mortality, LOS, and hospital charges following aortic and mitral valve replacements in the multivariable models that adjusted for the effect of race.

Discussion

The current study provides population based estimates of complications following aortic and mitral valve replacement surgeries and quantifies the impact of these complications on in- hospital mortality, LOS and hospital charges. The results clearly demonstrate that postoperative complications are prevalent and exert considerable influence on clinical outcomes. Over a third of patients undergoing these procedures experience complications following the surgery. Almost half of these are cardiac complications and about a quarter involve hemorrhage/hematoma/ seroma. Most of the studies that have examined complications following heart valve replacements are single center studies and mostly report on prosthesisrelated complications.5- 12-14 The current study reports on a wider range of possible postoperative complications, uses a nationally representative sample, and quantifies the economic impact of different types of complications.

TABLE V-Impact of complications on in-hospital mortality following aortic valve or mitral valve replacement (multivariable analysis).

Complications occurred more frequently in older patients after both surgeries, and in males after mitral valve replacement. As expected, patients with nonelective admissions were more likely to have complications. Of clinical note, patients undergoing repair with bioprosthetic valves were more prone to complications following both types of surgery than were patients undergoing mechanical valve replacement. In the current study, the odds of in-hospital mortality were higher for patients having replacements with bioprosthetic valves as compared to mechanical valves for both types of surgery in the bivariate analysis. However, type of valve replacement was not significantly associated with in-hospital mortality in the multivariable analysis thus suggesting that patient characteristics related to choice of prosthesis type may be significant confounders.

In addition to being prevalent, complications have serious consequences. In-hospital mortality rates, LOS, and hospital charges increase substantially with the occurrence of complications. Of the specific types of complications examined, almost all were associated with increased mortality rates, LOS, and hospital charges even after adjusting for patient and hospital characteristics. Of all the specific types of complications, patients with septicemia had the highest mortality rates, LOS and hospital charges. Previous studies have examined risk factors for mortality following heart valve replacement surgeries, and like this study have found that older age, female sex, emergent/urgent admissions, comorbid conditions (such as heart failure, chronic renal failure, and cerebrovascular disease), and tissue valve prosthesis are risk factors for mortality.2-4′ 6. 8 The current study goes beyond these previous investigations to demonstrate the significant impact of post- surgical complications on mortality, even after adjusting for these other contributing factors.

A large number of studies have found that hospital volume is associated with lower in-hospital mortality rates following complex surgical procedures.1 For aortic valve replacement surgeries, we found that there was an inverse relationship between hospital volume and in-hospital mortality even after adjusting for all other hospital and patient-related factors including complications. This may suggest that the positive effect of hospital volume operates through mechanisms other than the occurrence of complications, but this finding is too preliminary to draw conclusions other than to suggest that the role of complications in the relationship between post-surgical mortality and hospital volume should be further explored. Other hospital capacity characteristics such as bed size and teaching status were unrelated to occurrence of post-surgical complications or to outcomes. Several studies have examined predictors of outcomes following valve replacement surgeries using the Society of Thoracic Surgeons (STS) National Cardiac Surgery Database. J9-2i The STS National Cardiac Surgery Database has information on clinical parameters and the predictors of adverse outcomes as identified by studies using this database are more robust as compared to our study. However, it should be noted that the Nationwide Inpatient Sample is designed to be nationally representative of all nonfederal hospitals in the United States and our study adds value in providing nationally representative estimates of complications and quantifying the economic impact of complications on a national scale.

The current study has several limitations. The study uses administrative discharge data and the dataset does not contain information regarding severity of primary or secondary diagnoses. Unlike the Society of Thoracic Surgeons National Cardiac Surgery Database, the Nationwide Inpatient Sample does not have information on clinical parameters such as severity of left ventricular dysfunction, hemodynamic severity of valve lesions, and information on re-operations which precludes us from conducting a more comprehensive risk adjustment.22- 23 Another limitation is that the current analyses only use in-hospital mortality to assess short- term mortality following aortic and mitral valve replacement surgeries. It is quite possible that patients with complications have higher post-discharge mortality rates. So the estimated risk adjusted in-hospital mortality rates in this study may underestimate the true mortality rates for patients with complications. Finally, ICD-9-CM codes were used for identifying complications. Datasets created from hospital discharge abstracts may be prone to coding errors and biases, although examination of ICD-9-CM codes used to identify complications have been corroborated by review of medical records in 89% of surgical cases.24

Conclusions

Complications are prevalent and exert a considerable influence on clinical outcomes following aortic and mitral valve replacement surgeries. In-hospital mortality rates, LOS, and hospital charges increase in the presence of complications and increase further if more than one type of complication occurs. Our study results represent usual practices and thus could serve as benchmarks for future studies examining complication rates following heart valve replacement surgeries. Clearly, development of complications more than doubles the risk of in-hospital mortality. Thus quality initiatives should focus on reducing occurrence of complications and improving processes of care that would enable complications to be better resolved if they do occur.

References

1. Birkmeyer JD, Siewers AE, Finlayson EV, Stukel TA, Lucas FL, Batista I et al. Hospital volume and surgical mortality in the United States. N Engl J Med 2002;346: 1128-37.

2. Astor BC, Kaczmarek RG, Hefflin B, Daley WR. Mortality after aortic valve replacement: results from a nationally representative database. Ann Thorac Surg 2000;70:1939-45.

3. Nowicki ER, Birkmeyer NJ, Weintraub RW, Leavitt BJ, Sanders JH, Dacey LJ et al. Multivariable prediction of in-hospital mortality associated with aortic and mitral valve surgery in Northern New England. Ann Thorac Surg 2004;77:1966-77.

4. Birkmeyer NJ, Martin CA, Morton JR, Leavitt BJ, Lahey SJ, Charlesworth DC et al. Decreasing mortality for aortic and mitral valve surgery in Northern New England. Northern New England Cardiovascular Disease Study Group. Ann Thorac Surg 2000;70:432-7.

5. Blackstone EH, Cosgrove DM, Jamieson WR, Birkmeyer NJ, Lemmer JH, Miller DCJr et al. Prosthesis size and long-term survival after aortic valve replacement. J Thorac Cardiovasc Surg 2003;126:783-96.

6. Bloomstein LZ, Gielchinsky I, Bernstein AD, Parsonnet V, Saunders C, Karanam R et al. Aortic valve replacement in geriatric patients: determinants of in-hospital mortality. Ann Thorac Surg 2001;71:597-600.

7. Collait F, Feier H, Kerbaul F, Mouly-Bandini A, Riberi A, Mesana TG et al. Valvular surgery in octogenarians: operative risks factors, evaluation of Euroscore and long term results. Eur J Cardiothorac Surg 2005;27:276-80.

8. Florath I, Rosendahl UP, Mortasawi A1 Bauer SF, Dalladaku F, Ennker IC et al. Current determinants of operative mortality in 1400 patients requiring aortic valve replacement. Ann Thorac Surg 2003;76:75-83.

9. Hassan A, Quan H, Newman A, Ghali WA, Hirsch GM. Outcomes after aortic and mitral valve replacement surgery in Canada: 1994/ 95 to 1999/2000. CanJ Cardiol 2004;20:155-63.

10. Verheul HA, van den Brink RB, Bouma BJ, Hoedemaker G, Moulijn AC, Dekker E et al. Analysis of risk factors for excess mortality after aortic valve replacement. J Am Coll Cardiol 1995;26:1280-6.

11. Akins CW, Hilgenberg AD, Vlahakes GJ, Madsen JC, MacGillivray TE. Aortic valve replacement in patients with previous cardiac surgery. J Card Surg 2004;19:308-12.

12. Concha M, Aranda PJ, Casares J, Merino C, Alados P, Munoz I et al. Prospective evaluation of aortic valve replacement in young adults and middle-aged patients: mechanical prosthesis versus pulmonary autograft. J Heart Valve Dis 2005;14:40-6.

13. Jamieson WR, von Iipinski O, Miyagishima RT, Burr LH, Janusz MT, Ling H et al. Performance of bioprostheses and mechanical prostheses assessed by composites of valve-related complications to 15 years after mitral valve replacement. J Thorac Cardiovasc Surg 2005;129:1301-8.

14. Kvidal P, Bergstrom R, Malm T, Stahle E. Long-term follow-up of morbidity and mortality after aortic valve replacement with a mechanical valve prosthesis. Eur Heart J 2000;21:1099-111.

15. HCUP Databases. Healthcare Cost and Utilization Project (HCUP). May 2006, Agency for Healthcare Research and Quality, Rockville, MD. www.hcup-us.ahrq.gov/databases.jsp

16. Bureau of Labor Statistics. Inflation rate for Inpatient Services. http://www.bls.gov/cpi/. Date of access is 06/23/2005.

17. Elixhauser A, Steiner C, Harris DR, Coffey RM. Comorbidity measures for use with administrative data. Med Care 1998;36:8-27.

18. Hanley JA, Negassa A, Edwardes MD, Forrester JE. Statistical analysis of correlated data using generalized estimating equations: an orientation. AmJ Epidemiol 2003;157:364-75.

19. Rankin JS, Hammill BG, Ferguson TB, Glower DDJr, O’Brien SM, DeLong ER et al. Determinants of operative mortality in valvular heart surgery. J Thorac Cardiovasc Surg 2006;131:547-57.

20. Taylor NE, O’Brien S, Edwards FH, Peterson ED, Bridges CR. Relationship between race and mortality and morbidity after valve replacement surgery. Circulation 2005;111:1305-12.

21. Mehta RH, Eagle KA, Coombs LP, Peterson ED, Edwards FH, Pagani FD et al. Influence of age on outcomes in patients undergoing mitral valve replacement. Ann Thorac Surg 2002;74:1459-67.

22. Ferguson TB, Dziuban SWJr, Edwards FHJr, Eiken MC, Shroyer AL, Pairolero PC et al. The STS National Database: current changes and challenges for the new millennium. Committee to Establish a National Database in Cardiothoracic Surgery, The Society of Thoracic Surgeons. Ann Thorac Surg 2000;69:680-91.

23. Welke KF, Ferguson TB, Coombs LPJr, Dokholyan RS, Murray CJ, Schrader MA et al. Validity of the Society of Thoracic Surgeons National Adult Cardiac Surgery Database. Ann Thorac Surg 2004;77:1137-9.

24. Lawthers AG, McCarthy EP, Davis RB, Peterson LE, Palmer RH, lezzoni LI. Identification of in-hospital complications from claims

V. ALLAREDDY1, M. M. WARD1, J. W. ELY2, V. ALLAREDDY3, J. LEVETT4

1 Department of Health Management and Policy, College

of Public Health, The University of Iowa, E107 General

Hospital, Iowa City, IA, USA

2 Department of Family Medicine, The University of Iowa

University of Iowa Hospitals and Clinics

Iowa City, IA, USA

3 Departments of Internal Medicine and Pediatrics

MetroHealth Medical Center, Cleveland, OH, USA

4 Physicians’ Clinic of Iowa, Cedar Rapids, IA, USA

Acknowledgements.-This work was supported in part by Grant * HS015009 from the Agency for Healthcare Research and Quality.

Received on July 5, 2006.

Accepted for publication on April 27, 2007.

Address reprint requests to: Ward M. M., Ph.D., Professor and Associate Head, Department of Health Management and Policy, Director, Center for Health Policy and Research, College of Public Health, University of Iowa, 200 Hawkins Dr. E210GH, Iowa City, Iowa 52242-1008, USA. E-mail: [email protected]

Comments

comments