Best Evidence Topics
  • Send this BET as an Email
  • Make a Comment on this BET

What are the long-term results of cardiac valve replacements in left sided endocarditis with a history of i.v. drug abuse?

Three Part Question

In [i.v. drug users with left sided endocardits] does a [tissue or mechanical heart valve] affect [long-term mortality]?

Clinical Scenario

You are about to perform an urgent aortic valve replacement in a 32-year-old male who has been treated for endocarditis with destruction of the native aortic valve, leading to congestive heart failure. The patient has been an i.v. drug user (IVDU) for several years. He has a supportive family and has already been accepted into a methadone program in your community. In order to decide whether you should insert a mechanical heart valve or a tissue valve, you would like to learn about the long-term results of valve replacement in this clinical setting.

Search Strategy

Medline 1966–March 2006 using the OVID interface
[exp Substance Abuse, Intravenous/OR exp Substance Related-Disorders/OR drug OR drug user$.mp OR addict$.mp] AND [exp Endocarditis, Bacterial OR or exp Endocarditis/or exp Endocarditis, Subacute Bacterial/]. In addition, the search was re-performed omitting the term [exp heart valve prosthesis/].

Search Outcome

A total of 286 publications were reviewed and 9 papers describing cohort studies on more than 10 patients were selected as representing the best evidence on the subject

Relevant Paper(s)

Author, date and country Patient group Study type (level of evidence) Outcomes Key results Study Weaknesses
Mathew et al,
80 IVDU patients with endocarditis, 67 left sided endocarditis 27 mechanical valves and 38 bioprosthesis 1982-91Retrospective cohort study (level 2b)30 day mortality and follow-up events30 day mortality 7.6%, late mortality 19% (cardiovascular cause in 69%), survival at 5 years 74%, 10% recurrence of endocarditisMortality includes all patients (also right sided endocarditis) unknown which type of prosthesis had better results
Mestres et al,
Consecutive series of 31 HIV+cardiac valve operations 15 patients with left sided endocarditis 1985-2002Retrospective cohort study (level 2b)Early mortality, major morbidity and survival23% hospital mortality, 3 year survival 35% (overdoese commonest cause of death) 1 recurrent endocarditisAll patients were HIV1 positive
Chong et al,
Consecutive series of 20 HIV+patients with IVDU and left sided endocarditis 1990-1999Retrospective cohort study (level 2b)Early mortality, major morbidity and survivalNo hospital deaths, 50% mortality at 4 years. 25% recurrent endocarditis. All patients who continued IVDU died within 10 yearsNo data on prosthetic type
Arbulu et al,
Consecutive series of 280 IVDU, 110 valve operations, 60 for left sided endocarditis 1968-84Retrospective cohort study (level 4)Mortality, continued IVDU15% early mortality and 21% late mortality for all operations. 70 abandoned IVDU, all are long-term survivors. 23 hospital survivors returned to IVDU and all died
Levitsky et al,
Consecutive series of 37 IVDUs, 28 with left sided endocarditis 1976-81Retrospective cohort study (level 4)Early mortality, major morbidity and late mortality3.5 % early mortality, 37% late mortality, 10 patients with recurrent endocarditis all with continued IVDUUnstructured follow-up
Mammana et al,
Series of 18 IVDU with left sided endocarditis 1976-79Retrospective cohort study (level 4)Early and late mortality, complications30 day mortality 11%, late mortality 44%. All deaths occurred within 9 months due to persistent or recurrent endocarditisUnstructured follow-up The only publication focussing on left sided endocarditis (Patients are included in reference Levitsky, 1982)
Hiratzka et al,
Study on 32 consecutive patients with endocarditis, 16 IVDU with left sided endocarditis 1972-77Retrospective cohort study (level 4)30 day mortality, late mortality, postoperative complications30 day mortality, 6% in IVDU, 7% in non-IVDU

Late mortality 37% in IVDU, 7% in non-IVDU. 5 of 7 deaths in IVDU group due to continued IV drug use
Unstructured follow-up, type of valves implanted unknown
Hubbel et al,
79 IVDU's with endocarditis, 24 operated for left sided endocarditis 1965-76Retrospective cohort study (level 4)Clinical follow-up12% perioperative deaths, 11 patients died during follow-up, 24% reinfectionsUnstructured follow-up


The largest study was published by Mathew et al. Out of 80 surgical patients included in the series, 65 had a left sided valve replacement (27 mechanical prosthesis and 38 bioprosthesis). The 30-day mortality was 7.6% and late mortality was 19% for the whole series. Survival at 5 years was 70%. Thirty percent developed at least one major cardiovascular event. The results for patients with left sided endocarditis were not given separately. The most contemporary series was published by Mestres et al. summarizing the experience with 31 HIV-positive patients. Twenty of them had endocarditis and were IVDUs (15 left sided). Bioprosthesis and mechanical prosthesis were inserted. The hospital mortality in the endocarditis group was 23.8% and 3-year survival was 35%. The commonest cause of death was a drug overdose. Only one patient died from recurrent endocarditis. The study by Chong et al. includes 22 HIV-positive patients, 20 were IVDUs and suffered from left sided endocarditis. The hospital moratility was 0%, late mortality was 50% at 4 years and there was a 25% recurrence rate for endocarditis. All patients who continued on IVDU were dead within 10 years. None of those who discontinued IVDU developed recurrent endocarditis. The authors recommend using mechanical prosthesis because it prevents reoperations in those who discontinue IVDU. The type of prosthesis has no influence on survival in those who continue IVDU. The series published by Arbulu et al. covers a time span of 17 years and comprises a large number of patients with endocarditis who were IVDUs. The paper includes the results on 110 patients who were operated on for endocarditis, 60 had left sided endocarditis. The early results for the whole group was a 15% mortality and a 21% late mortality. A separate analysis for patients with left sided endocarditis is not given. One hundred and seventy-five percent of surviving patients abandoned IVDU and were apparently long-term survivors. All 23 patients who continued using i.v. drugs eventually died. Eleven of these patients were reoperated for recurrent endocarditis and the authors conclude that reoperation in this patient group is contraindicated. The reports of Levitsky et al. and Mammana et al. cover different aspects of surgery for endocarditis in IVDU using data from the same institution. Levitsky et al. report on 28 patients with left sided endocarditis who were subjected to bioprosthetic valve replacement. Thirty-day mortality was 3.5% and long term mortality was 37%. Ten patients required reoperation for recurrent endocarditis. All of these patients had continued IVDU. Mamman's report covers exclusively left sided endocarditis. The 30-day mortality was 11% and overall mortality 50%. All deaths occurred within 9 months of the initial operation, mostly due to persistent sepsis. Fifty-six percent of survivors admitted to continued drug addiction. Hiratzka et al. in 1979 reported a retrospective cohort study on 32 consecutive patients undergoing surgery for endocarditis over a six year period including 16 patients with IVDU. Thirty-day mortality was comparable in IVDU and non-IVDU (6% vs. 7%). Seven of the 17 hospital survivors (41%) with IVDU died during follow-up, 5 of them due to continued IVDU (3 patients died of an overdose and 2 due to recurrent infection). In non-IVDU only one patient died during follow up (8%). Seven patients developed prosthetic endocarditis, 6 of them in the IVDU group. All patients in the IVDU group returned to their addiction. Ninety-seven episodes of endocarditis in 79 patients with IVDU were retrospectively reviewed by Hubbel et al. A cardiac valve replacement was performed in 24 patients with left sided endocarditis. The perioperative mortality was 16%. Twenty-four percent of the surgical patients developed recurrent endocarditis and 11 patients died during follow up. Thus, nearly 60% of the surgical patients with left sided endocarditis were dead at a maximum follow up of 29 months. Frater published a large series on 57 patients (43 left sided endocarditis). The study was excluded because 37% of patients were lost to follow up.

Clinical Bottom Line

Cardiac valve replacement in IVDUs with left sided endocarditis carries a substantial long-term mortality. Continued IVDU leading to recurrent endocarditis and death from drug overdose are the commonest causes of morbidity and mortality. HIV infection seems to have little influence on survival according to the literature. Patients who discontinue IVDU have a favorable prognosis and implantation of a mechanical prosthesis is warranted. Every effort should be made to prevent these patients from returning to IVDU. We recommend consulting experts locally, since excellent results with abstinence in up to 80% of addicts may be in some methadone substitution programs [Kornor ].


  1. Mathew J, Abreo G, Namburi K, Narra L, Franklin C. Results of surgical treatment for infective endocarditis in intravenous drug users. Chest 1995; 108:73–77.
  2. Mestres CA, Chuquiure JE, Claramonte X, Munoz J, Benito N, Castro MA, Pomar JL, Miro JM. Long-term results after cardiac surgery in patients infected with the human immunodeficiency virus type-1 (HIV-1). Eur J Cardiothorac Surg 2003; 23:1007–1016.
  3. Chong T, Alejo DE, Greene PS, Redmond JM, Sussman MS, Baumgartner WA, Cameron DE. Cardiac valve replacement in human immunodeficiency virus-infected patients. Ann Thorac Surg 2003; 76:478–480.
  4. Arbulu A, Asfaw I. Management of infective endocarditis: seventeen years' experience. Ann Thorac Surg 1987; 43:144–149.
  5. Levitsky S, Mammana RB, Silverman NA, Weber F, Hiro S, Wright RN. Acute endocarditis in drug addicts: surgical treatment for gram-negative sepsis. Circulation 1982; 66:Suppl I, I135–138.
  6. Mammana RB, Levitsky S, Sernaque D, Beckman CB, Silverman NA. Valve replacement for left-sided endocarditis in drug addicts. Ann Thorac Surg 1983; 35:436–441.
  7. Hiratzka LF, Nelson RJ, Oliver CB, Jengo JA. Operative experience with infective endocarditis. Drug users compared with non-drug users. J Thorac Cardiovasc Surg 1979; 77:355–361.
  8. Hubbell G, Cheitlin MD, Rapaport E. Presentation, management, and follow-up evaluation of infective endocarditis in drug addicts. Am Heart J 1981; 102:85–94.
  9. Frater RW. Surgical management of endocarditis in drug addicts and long-term results. J Card Surg 1990; 5:63–67.
  10. Kornor H, Waal H. From opioid maintenance to abstinence: a literature review. Drug Alcohol Rev 2005; 24:267–274.