Left ventricular assist devices (LVADs) have become an established therapy for patients with end-stage heart failure.1 Although these devices can bridge patients to cardiac transplantation,2 they can also be used as permanent, long-term therapy.3 As the utilization of these devices continues to rise, and survival continues to increase,4 the number of LVAD patients with noncardiac surgical disease will increase.
Left ventricular assist device patients can present a unique challenge to caretakers. Anesthesiologists, operating room staff, and surgeons may not be familiar with the management of these devices.5 Previous studies have suggested that noncardiac surgical procedures (NCSs) can safely be carried out in patients supported with LVAD.6–9 However, few studies investigated the need for NCSs during the same admission as LVAD implantation. In addition, no study to date has examined this topic in a large, national database. The goal of our study was to define the type of NCSs required in patients during the same hospital admission as LVAD implantation. In addition, we attempt to measure the outcomes of patients undergoing NCSs after LVAD implantation. Finally, we attempt to determine the factors that affect outcomes in LVAD patients requiring NCSs.
Patients and Methods
Data Source and Study Population
The National Inpatient Sample (NIS) database was queried from 2007 to 2010. The NIS database consists of data from approximately eight million hospital stays each year and represents a stratified sample of 20% of nonfederal hospitals in the United States. The NIS database is the largest all-payer inpatient health-care database in the United States.10 International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM) procedure codes for permanent, implantable LVAD (37.66) were used to identify all adult patients requiring intracorporeal, LVAD support. Patients undergoing NCSs after the implantation of LVAD were compared with other LVAD patients not requiring procedures. Types of procedures performed as well as diagnostic and comorbidity data were obtained by ICD-9 diagnosis codes as established in previous studies.11,12 The NIS database does not include information on time elapsed from LVAD implantation and performance of NCS.
The primary outcome measured was inpatient mortality after NCS. Secondary outcomes included perioperative complications using ICD-9 codes representing wound infections, pneumonia, urinary tract infection (UTI), bleeding complications, acute renal failure, pulmonary embolus/deep vein thrombosis (DVT), and sepsis as established in previous studies.11,12 The length of stay and total hospital costs were also examined.
Student’s t-test and chi-squared test were used to examine continuous and categorical variables where appropriate. Continuous variables are presented as mean ± standard deviation when data were normally distributed. Categorical variables in Table 1 list NCS tallied as multiple events per patient. Categorical variables in Tables 2 and 3 are reported as percentages of the total number of data points available for that field for each index patient. Multiple variable logistic regression analysis using clinically relevant variables was carried out to examine variables associated with mortality. Variables included in the analysis were age, sex, Charlson comorbidity index, insurance status, hospital bed size capacity, race, year of admission, and income. These covariates have been shown to be clinically relevant in previous studies.11,12 Subset analysis was carried out to examine general surgery procedure, thoracic surgery procedures, and NCSs in the same multivariate analysis. A p value less than 0.05 was considered statistically significant. Data were analyzed using SAS statistical software version 9.2 (SAS Institute, Cary, NC).
There were a total of 1,397 patients implanted with an LVAD during the study period. Of these, 298 (21.3%) required 459 NCSs for an average of 1.54 procedures per patient. A total of 1,099 (78.7%) patients implanted with an LVAD did not require an NCS.
Types of Procedures Required
A summary of the types of NCSs required by patients implanted with LVAD is shown in Table 1. The most common type of procedure required was general surgical in nature (n = 153, 33.3%). The most common types of general surgical procedures required were abdominal exploration/bowel procedures (n = 76, 16.6%) and wound debridement/soft tissue surgery (n = 44, 9.6%). The second most common category of NCSs was noncardiac thoracic procedures (n = 141, 30.7%). The most common thoracic procedures were exploratory thoracotomy/thoracoscopy (n = 71, 15.5%) and chest wall procedures (n = 71, 15.5%). The third most common category was vascular/extremity surgical procedures (n = 140, 30.5%). The most common vascular/extremity procedures were the repair of vessels (n = 44, 9.6%), major vascular bypass (n = 34, 7.4%), and angiography/endovascular intervention (n = 32, 7.0%).
Baseline Patient Characteristics
A comparison of baseline patient characteristics is shown in Table 2. The two groups were not significantly different with respect to age, sex, or Charlson comorbidity index. In addition, the patients were not significantly different with respect to the number of elective admissions, insurance status, and ethnicity. Patients requiring vascular procedures were more likely to require extracorporeal membrane oxygenation (ECMO) support as a bridge to LVAD (n = 18, 6.1%) when compared with other patients requiring nonvascular procedures (n = 7, 2.4%) (p < 0.001). Patients requiring vascular procedures were not more likely to have an intra-aortic balloon pump before LVAD (n = 5, 2.6%) when compared with LVAD patients having nonvascular procedures (n = 1, 0.3%) (p = 0.32).
A comparison of inpatient complications is shown in Table 3. The length of stay was not significantly different when comparing the NCS cohort with LVAD patients that did not require surgery (42.4 vs. 38.2 days, p = 0.09). The noncardiac surgery group was more likely to develop surgical site infections (9.1 vs. 4.6%, p = 0.004), bleeding complications (44.0 vs. 24.8%, p < 0.001), or any complication (87.2 vs. 82.0, p = 0.001). The incidence of pneumonia, UTI, acute renal failure, pulmonary embolus/DVT, or sepsis was not significantly different. Total hospital charges were not significantly different when comparing the two groups ($895,655.00 vs. $787,964.00, p = 0.09).
The noncardiac surgery cohort did not have a higher incidence of inpatient mortality (22.8 vs. 17.9%, p = 0.10). Results of the multiple variable analysis are shown in Table 4. Undergoing NCSs was not associated with mortality (odds ratio [OR]: 1.45, 95% confidence interval [CI]: 0.95–2.20, p = 0.08). The only variable associated with mortality included a small hospital bed size capacity (OR: 2.59, 95% CI: 1.98–3.39, p < 0.001). Variables associated with survival included more recent hospital admissions in 2009 (OR: 0.49, 95% CI: 0.25–0.61, p = 0.02) and 2010 (OR: 0.32, 95% CI: 0.20–0.51, p < 0.001).
Subset analysis was carried out in the same multiple variable analysis to determine whether the different categories of NCSs were associated with mortality. General surgery procedures (hazards ratio (HR): 1.05, 95% CI: 0.95–1.16, p = 0.41), noncardiac thoracic procedures (HR: 1.82, 95% CI: 0.45–7.39, p = 0.46), and vascular procedures (HR: 2.02, 95% CI: 0.51–8.02), p = 0.23) were not independently associated with mortality.
As the survival in LVAD patients improves,13 and number of patients implanted with these devices continues to increase,4 the number of LVAD patients with noncardiac surgical problems will rise.14 Previous research has shown that NCSs are safe in patients with LVAD,15 but no study to our knowledge has examined this topic in a large, national database. At present, the largest study to examine this topic consisted of 47 patients with LVAD undergoing NCSs.16 Our study investigated the types of procedures required during the same hospitalization as LVAD implantation and examined variables that may affect outcomes in these patients. Left ventricular assist device patients requiring NCS have higher bleeding and infectious complications than LVAD patients not requiring NCS. In addition, the complication and mortality rate of LVAD patients requiring NCS appear to be higher than non-LVAD patients undergoing similar procedures.
Small sample sizes have limited the ability of prior studies to characterize the types of NCSs that are common in LVAD patients.5,8,9,15–18 Our study demonstrates that the most common types of procedures required by these patients include general surgical, thoracic, and vascular/extremity. Among patients requiring general surgery procedures, the most common are abdominal exploration and/or bowel procedures. Previous studies have documented that abdominal pathology does occur among LVAD patients.16–19 Gastrointestinal bleeding is also common among LVAD patients and can result in the need for abdominal exploration.20 Wound debridements and soft tissue surgeries were also common among general surgical procedures. Surgical site infections and soft tissue pathology requiring operation in the LVAD population have not been established in previous studies and needs further investigation.
Noncardiac thoracic procedures and vascular/extremity surgeries were also common among patients implanted with LVAD. This finding has not been documented in previous studies. Patients requiring vascular procedures were more likely to have ECMO before LVAD implantation. This is corroborated by other studies, as the use of ECMO is a known risk factor for vascular complications.21 However, only a small number of patients requiring vascular procedures (6.1%) were on ECMO before LVAD implantation, which suggests that vascular complications cannot be solely attributed to the use of ECMO. The most common noncardiac, thoracic procedures were exploratory thoracotomy/thoracoscopy and chest wall procedures. Unfortunately, the diagnosis leading to these procedures cannot be elicited from the NIS database. It is possible that many of these patients have coexisting lung disease that required open or thoracoscopic lung biopsy over the course of their hospitalization. Further studies are necessary to examine why noncardiac, thoracic procedures are so common in LVAD patients. Future investigation on the requirement for vascular procedures in LVAD patients is also needed. Our study noted that the most common vascular procedures were the repair of vessels, major vascular bypass, and angiography/endovascular intervention. Left ventricular assist device patients are known to be at increased risk of thromboembolic complications,22–24 and this may lead to increased need for vascular/extremity procedures.
Although NCSs in LVAD patients had an OR of 1.45, this did not reach statistical significance (p = 0.08). This was found to be true even when looking specifically at general surgery procedures, noncardiac thoracic procedures, and vascular procedures. This finding is consistent with previous studies that have shown that both emergent and elective procedures can be carried out in LVAD patients without adverse outcomes.5,8,9,15–19 Our study was unique in that it examined LVAD patients that had NCSs during the same admission as LVAD implantation. A potential limitation is that this study was underpowered and did not have enough deaths to detect an association with mortality. Left ventricular assist device patients requiring NCS had a relatively high mortality rate of 22.8%, which is likely higher than non-LVAD patients requiring similar procedures.
Other variables that affect survival include smaller bed size capacity and implantation of LVAD in more recent years. Although the impact of hospital bed size on LVAD outcomes has not been studied, other institutional factors, such as postoperative care and patient selection, may influence outcomes.25 The volume of operation is also known to affect operative outcomes.26 Smaller hospital bed may be a marker of other institutional factors, such as volume or patient selection, which can influence outcomes in these patients. In addition, larger hospital bed size may represent institutions that are actively implanting and managing patients with LVADs. This may explain why outcomes are better at these hospitals, as experts are available to help manage complications that are unique to this patient population, such as pump thrombosis and gastrointestinal bleeds. This suggests that LVAD patients undergoing even low-risk surgeries should be cared for at such institutions with mechanical circulatory support specialists. The increased mortality seen in smaller hospitals may represent nonmechanical circulatory support centers doing NCS in LVAD patients
In this study, hospital admission in more recent years was associated with survival. This finding is consistent with existing studies that have shown improved outcomes in LVAD patients as experience with these devices continues to increase.27,28 The reasons for improved outcomes over time are likely multifactorial and include better surgical technique,24 improved patient management,28,29 and a transition from pulsatile flow devices to continuous flow devices.22,30,31
Although survival was not affected by the need for NCSs, patients that underwent these procedures had higher morbidity. They had a greater number of wound infections, bleeding complications, and any complication. The higher number of wound complications is not surprising as the noncardiac surgery cohort had a greater number of surgical wounds. A higher number of bleeding complications are also to be expected for the same reason. In addition, patients with LVAD are routinely anticoagulated and on Aspirin caused by the risk for thromboembolic complications and pump thrombosis. As a result, surgeons must carefully weigh the risks of having a bleeding complication against inducing a thromboembolic event. Prior studies have shown that as many as 25.5% of LVAD patients can have bleeding complications when requiring NCSs. However, our study demonstrates that this problem is even more profound, with 44.0% of LVAD patients undergoing NCSs having bleeding complications. This shows that there is much room for improvement in the perioperative management of anticoagulation in these patients and that further studies are needed. Unfortunately, no clear guidelines exist in how to manage anticoagulation in the perioperative period for LVAD patients. Previous studies have shown that anticoagulation goals can be safely lowered in LVAD patients for a short period of time.19,32 However, antiplatelet therapy should be continued throughout the perioperative period due to their benefits in LVAD patients.19,33,34 Further studies are needed to decrease postoperative bleeding complications in these patients.
Although the incidence of UTI, renal failure, pneumonia, and sepsis were not significantly different when comparing the NCS and non-NCS groups, LVAD patients requiring NCS have a high degree of postoperative complications when compared to previously published reports. The incidence of postoperative UTI has been estimated to be 10%.35 By comparison, this study showed that LVAD patients requiring NCS had UTI in 14.1% of cases. In addition, the incidence of postoperative pneumonia was found to be 18.5% in this study, much higher than estimated rates of 3.3–8.1%, depending on the type of surgery.36 In addition, the incidence of postoperative sepsis has been estimated to be as low as 0.9%,37 but found to be 19.1% in LVAD patients undergoing NCS. The incidence of postoperative renal failure in LVAD patients undergoing NCS was 53.4%, whereas it has been estimated to be as high as 13% in previously published reports.38
This study was not without limitations. The NIS database is an administrative database, therefore; there are several variables that could impact outcomes but were not available for analysis. These potentially impactful factors include pre-existing surgical conditions, anticoagulation strategy, and the type of LVAD utilized. In addition, we could not obtain detailed information on the surgical procedures that each patient required or the complications that occurred during their hospital stay as these are obtained by ICD-9 and procedure codes. Finally, the database is limited to inpatient outcomes so we could not examine the long-term morbidity and mortality.
In conclusion, patients implanted with LVAD commonly require NCSs. The most common types of procedures are general surgical, with abdominal exploration being most common. In addition, noncardiac thoracic surgeries and vascular surgeries are common. The requirement of NCSs does not result in increased mortality risk; however, it does result in increased morbidity. In addition, morbidity and mortality in these patients appear to be higher than those of the general population requiring similar procedures. Finally, LVAD patients undergoing even low-risk NCS should be cared at mechanical circulatory support specializing centers with treating surgeons and LVAD specialists.
1. Slaughter MS, Rogers JG, Milano CA, et al: Advanced heart failure treated with continuous-flow left ventricular assist device. N Engl J Med 2009.361: 2241–2251.
2. Miller LW: Patient selection for the use of ventricular assist devices as a bridge to transplantation. Ann Thorac Surg 2003.75(6 suppl): S66–S71.
3. Lietz K, Long JW, Kfoury AG, et al: Outcomes of left ventricular assist device implantation as destination therapy in the post-REMATCH era. Circulation 2007.116: 497–505.
4. Lampropulos JF, Kim N, Wang Y, et al: Trends in left ventricular assist device use and outcomes among Medicare beneficiaries, 2004–2011. Open Heart 2014.1: e000109.
5. Morgan JA, Paone G, Nemeh HW, et al: Non-cardiac surgery in patients on long-term left ventricular assist device support. J Heart Lung Transplant 2012.31: 757–763.
6. Brown JB, Hallinan WM, Massey HT, et al: Does the need for noncardiac surgery during ventricular assist device therapy impact clinical outcome? Surgery 2009.146: 627–634.
7. Eckhauser AE, Melvin WV, Sharp KW: Management of general surgical problems in patients with left ventricular assist devices. Am Surg 2006.72: 158–161.
8. Garatti A, Bruschi G, Colombo T, et al: Noncardiac surgical procedures in patient supported with long-term implantable left ventricular assist device. Am J Surg 2009.197: 710–714.
9. Goldstein DJ, Mullis SL, Delphin ES, et al: Noncardiac surgery in long-term implantable left ventricular assist-device recipients. Ann Surg 1995.222: 203–207.
11. Iezzoni LI, Daley J, Heeren T, et al: Identifying complications of care using administrative data. Med Care 1994.32: 700–715.
12. Kilic A, Sheer A, Shah AS, Russell SD, Gourin CG, Lidor AO: Outcomes of cholecystectomy in US heart transplant recipients. Ann Surg 2013.258: 312–317.
13. Kirklin JK, Naftel DC, Kormos RL, et al: Fifth INTERMACS annual report: risk factor analysis from more than 6,000 mechanical circulatory support patients. J Heart Lung Transplant 2013.32: 141–156.
14. McKellar SH, Morris DS, Mauermann WJ, Park SJ, Zietlow SP: Evolution of general surgical problems in patients with left ventricular assist devices. Surgery 2012.152: 896–902.
15. Ahmed M, Le H, Aranda JM Jr, Klodell CT: Elective noncardiac surgery in patients with left ventricular assist devices. J Card Surg 2012.27: 639–642.
16. Arnaoutakis GJ, Bittle GJ, Allen JG, et al: General and acute care surgical procedures in patients with left ventricular assist devices. World J Surg 2014.38: 765–773.
17. Stehlik J, Nelson DM, Kfoury AG, et al: Outcome of noncardiac surgery in patients with ventricular assist devices. Am J Cardiol 2009.103: 709–712.
18. Schmid C, Wilhelm M, Dietl KH, Schmidt C, Hammel D, Scheld HH: Noncardiac surgery in patients with left ventricular assist devices. Surgery 2001.129: 440–444.
19. Taghavi S, Beyer C, Vora H, et al: Noncardiac surgery in patients on mechanical circulatory support. ASAIO J 2014.60: 670–674.
20. Crow S, John R, Boyle A, et al: Gastrointestinal bleeding rates in recipients of nonpulsatile and pulsatile left ventricular assist devices. J Thorac Cardiovasc Surg 2009.137: 208–215.
21. Taghavi S, Jayarajan SN, Mangi AA, et al: Examining Noncardiac Surgical Procedures in Patients on Extracorporeal Membrane Oxygenation. ASAIO J 2015.61: 520–525.
22. John R, Kamdar F, Liao K, Colvin-Adams M, Boyle A, Joyce L: Improved survival and decreasing incidence of adverse events with the HeartMate II left ventricular assist device as bridge-to-transplant therapy. Ann Thorac Surg 2008.86: 1227–1234; discussion 1234.
23. John R, Kamdar F, Liao K, et al: Low thromboembolic risk for patients with the Heartmate II left ventricular assist device. J Thorac Cardiovasc Surg 2008.136: 1318–1323.
24. Taghavi S, Ward C, Jayarajan SN, Gaughan J, Wilson LM, Mangi AA: Surgical technique influences HeartMate II left ventricular assist device thrombosis. Ann Thorac Surg 2013.96: 1259–1265.
25. Lietz K, Long JW, Kfoury AG, et al: Impact of center volume on outcomes of left ventricular assist device implantation as destination therapy analysis of the thoratec HeartMate registry, 1998 to 2005. Circ Heart Fail 2: 3–10, 2009.
26. Birkmeyer JD, Siewers AE, Finlayson EV, et al: Hospital volume and surgical mortality in the United States. N Engl J Med 2002.346: 1128–1137.
27. Khazanie P, Hammill BG, Patel CB, et al: Trends in the use and outcomes of ventricular assist devices among medicare beneficiaries, 2006 through 2011. J Am Coll Cardiol 2014.63: 1395–1404.
28. Park SJ, Milano CA, Tatooles AJ, et al: Outcomes in advanced heart failure patients with left ventricular assist devices for destination therapy. Circ Heart Fail 2012.5: 241–48.
29. Slaughter MS, Pagani FD, Rogers JG, et al: Clinical management of continuous-flow left ventricular assist devices in advanced heart failure. J Heart Lung Transplant 2010.29: S1–S39.
30. Kirklin JK, Naftel DC, Kormos RL, et al: Fifth INTERMACS annual report: Risk factor analysis from more than 6,000 mechanical circulatory support patients. J Heart Lung Transplant 2013.32: 141–156.
31. Taghavi S, Jayarajan SN, Komaroff E, Mangi AA: Continuous flow left ventricular assist device technology has influenced wait times and affected donor allocation in cardiac transplantation. J Thorac Cardiovasc Surg 2014.147: 1966–1971, 1971.e1.
32. John R, Liao K, Lietz K, et al: Experience with the Levitronix CentriMag circulatory support system as a bridge to decision in patients with refractory acute cardiogenic shock and multisystem organ failure. J Thorac Cardiovasc Surg 2007.134: 351–358.
33. Attisani M, Centofanti P, La Torre M, Campanella A, Sansone F, Rinaldi M: Safety and effectiveness of low dosing of double antiplatelet therapy during long-term left ventricular support with the INCOR system. J Artif Organs 2010.13: 202–206.
34. Meuris B, Arnout J, Vlasselaers D, Schetz M, Meyns B: Long-term management of an implantable left ventricular assist device using low molecular weight heparin and antiplatelet therapy: a possible alternative to oral anticoagulants. Artif Organs 2007.31: 402–405.
35. Stéphan F, Sax H, Wachsmuth M, Hoffmeyer P, Clergue F, Pittet D: Reduction of urinary tract infection and antibiotic use after surgery: A controlled, prospective, before-after intervention study. Clin Infect Dis 2006.42: 1544–1551.
36. Horan T, Culver D, Gaynes R, Jarvis W, Edwards J, Casetta R: Nosocomial infections in surgical patients in the United States, January 1986–1992. Infect Control Hosp Epidemiol 1993.14: 73–80.
37. Bateman BT, Schmidt U, Berman MF, Bittner EA: Temporal trends in the epidemiology of severe postoperative sepsis after elective surgery: A large, nationwide sample. Anesthesiology 2010.112: 917–925.
38. Wijeysundera DN, Karkouti K, Beattie WS, Rao V, Ivanov J: Improving the identification of patients at risk of postoperative renal failure after cardiac surgery. Anesthesiology 2006.104: 65–72.