Saudi Journal of Kidney Diseases and Transplantation

: 2011  |  Volume : 22  |  Issue : 1  |  Page : 160--166

Infective endocarditis in chronic hemodialysis patients: Experience from Morocco

Dina Montasser1, Abdelali Bahadi2, Yassir Zajjari2, Mohamed Asserraji2, Ahmed Alayoude2, Omar Moujoud2, Toufik Aattif2, Moncef Kadiri2, Nadir Zemraoui2, Driss El Kabbaj2, Mohamed Hassani2, Mohamed Benyahia2, Mustapha El Allam2, Zouhir Oualim2, Ismail Akhmouch1,  
1 Agadir Hemodialysis Center, First Medical Center, Agadir, Morocco
2 Department of Nephrology, Dialysis and Renal Transplantation, Military Hospital Mohammed V, Rabat, Morocco

Correspondence Address:
Ismail Akhmouch
Service de Nephro-Hemodialyse, Etat Major General, Agadir


Since the 1960s, regular hemodialysis (HD) was recognized as a risk factor for the development of infective endocarditis (IE), particularly at vascular access sites. The present report describes our experience at the Etat Major General Agadir, Morocco, of taking care of IE in patients on regular dialysis. A retrospective analysis was made of five cases of IE in patients receiving re­gular HD having arteriovenous fistula as vascular access. They were sent from four private centers and admitted in our formation between January 2004 and March 2009. Infective endocarditis was detected after 34.5 months following initiation of dialysis. The causative organisms included Sta­phylococcus and Enterococcus in two cases each and negative blood culture in one case. A recent history of infection (<3 months) of the vascular access was found in three cases. Peripheric embolic phenomena were noted in two cases. A pre-existing heart disease was common and contributed to heart failure. Mortality was frequent due to valvular perforations and congestive heart failure, making the medical treatment alone unsatisfactory. Two patients survived and three of our patients received a prosthetic valve replacement, with a median survival after surgery of 10.3 months/person. The clinical diagnosis of infective endocarditis in regularly dialyzed patients remains difficult, with the presence of vascular calcification as a common risk factor. The vascular catheter infections are the cardinal gateway of pathogenic organisms, which are mainly Staphlococcus. The prognosis is bad and the mortality is significant, whereas medical and surgical treatments are often established in these patients who have many factors of comorbidity.

How to cite this article:
Montasser D, Bahadi A, Zajjari Y, Asserraji M, Alayoude A, Moujoud O, Aattif T, Kadiri M, Zemraoui N, El Kabbaj D, Hassani M, Benyahia M, El Allam M, Oualim Z, Akhmouch I. Infective endocarditis in chronic hemodialysis patients: Experience from Morocco.Saudi J Kidney Dis Transpl 2011;22:160-166

How to cite this URL:
Montasser D, Bahadi A, Zajjari Y, Asserraji M, Alayoude A, Moujoud O, Aattif T, Kadiri M, Zemraoui N, El Kabbaj D, Hassani M, Benyahia M, El Allam M, Oualim Z, Akhmouch I. Infective endocarditis in chronic hemodialysis patients: Experience from Morocco. Saudi J Kidney Dis Transpl [serial online] 2011 [cited 2023 Feb 4 ];22:160-166
Available from:

Full Text


Patients on chronic hemodialysis (HD) are particularly vulnerable to bacteremia and to infec­tive endocarditis (IE) due, on one hand, to the vascular access needed for dialysis (arterio­venous fistula or central catheter) and, on the other hand, increased susceptibility to infection caused by end-stage renal failure.

Antibiotic treatment should take into conside­ration the specificity of endocardial damage, its kinetics during the inter-dialysis and the dialysis periods and its toxicity. The seriousness of the disease makes it mandatory to choose the best antibiotic to minimize toxicity without compro­mising the efficacy of treatment. Also, a spe­cific preventive measure must be implemented.

The purpose of our study is to evaluate the clinical, bacteriological and therapeutic aspects of IE in patients on chronic HD.

 Patients and Methods

This is a retrospective study conducted at the Department of Nephrology, Hemodialysis and Transplantation in the Military Hospital Moha­mmed V in Rabat and the HD unit in the First Medical Center in Agadir. The study involved five hospitalized patients in our center between January 2004 and March 2009.

The parameters studied were collected from patient records, and included the following: de­mographic characteristics such as gender and year of birth, specific medical history including etiology of CKD, year of initiation of dialysis, presence of diabetes, date of occurrence of IE after starting HD, history of infection, history of cardiac involvement and details regarding vas­cular access.

Clinical criteria noted included presence of fever, cardiac involvement (new or changing murmurs, heart failure, weight gain, episodes of acute pulmonary edema (APO), cutaneous le­sions like Osler's nodes, Roth's spot, subungual "splinter" hemorrhages, cutaneous petechiae and clubbing as well as presence of hepatospleno­megaly.

Radiological tests were performed to aid in the diagnosis of IE as well as to identify the site of location of vegetations and the presence of associated cardiac lesions. Additionally, trans­thoracic and transesophageal echocardiographs were performed.

Bacteriological diagnosis was based on per­sistently positive blood cultures, defined as growth of a microorganism, consistent with IE from all three or a majority of four or more separate blood cultures, with the first and the last drawn at least one hour apart. Blood tests were performed for leukocyte count, hemo­globin, platelet count and C-reactive protein.

The common and specific characteristics of IE in our study patients as well as the evolution and survival following the treatment were studied.

A descriptive analysis was performed by the "SPSS" program in order to translate the qua­litative variables as percentage and quantitative variables as average.


Between January 2004 and March 2009, five patients, three females and two males, deve­loped endocarditis while on maintenance dia­lysis in our center. Their mean age was 48 years (range: 31-66 years). Infective endocarditis was detected after a mean of 35.4 months (4-102 months) after dialysis was initiated. Four cases occurred on a native valve and one case oc­curred on a prosthetic valve. Three patients had a history of repeated infection of the arterio­venous fistula less than three months earlier and two cases had infection of the femoral dual lu­men catheter. All our patients had a native arte­riovenous fistula in the left upper limb at ad­mission, of which three were radiocephalic and two were brachiocephalic. All study patients had valvular calcifications. The results are summa­rized in [Table 1].{Table 1}

The symptoms were dominated by fever found in four of five cases, a heart murmur that remained almost constant in all five, congestive heart failure in three patients at diagnosis of IE, clubbing that meant severe hypoxia in all five patients, embolic phenomenon that was seen in two of our patients in the form of liver infarc­tion and a splenic abscess and Osler nodes and vascular injury causing necrotic purpura affec­ting the lower limb in one patient each. Epi­sodes of APO between dialysis was present in all our study patients. None of the patients de­veloped neurological events. The data are sum­marized in [Table 2].

At least three blood cultures out of three were positive in four patients while one patient had negative blood cultures on all four occasions. The pathogens grown included Staphylococcus aureus and Enterococcus fecalis group D in two cases each, with one case having negative blood cultures. We noted an evident infectious and in­flammatory syndrome and severe anemia in the study patients (mean Hb 7.5 g/dL, range 6.2-10.4 g/dL) requiring blood transfusions. A positive positive cryoglobulinemia (not typed) was de­tected in one patient with necrotic purpura with negative hepatitis serology. Data are summa­rized in [Table 3].{Table 2}{Table 3}

The diagnosis was established by trans-thoracic (TTE) and trans-esophageal (TEE) echocardio­graph, except in two cases who did not tolerate the procedure. One patient had endocarditis of the aortic valve, one other of the mitral and aortic valves, two cases had mitral endocarditis and one case had involvement of the mitral and aortic prosthetic valves. The associated cardiac events included aortic and mitral insufficiency in three cases, perforation of the aortic and mit­ral valves in two cases and a trigonal mitral and aortic abscess in one case [Table 4].{Table 4}

Episodes of APO were treated symptomati­cally with oxygen and daily sessions of HD and ultrafiltration to adjust the dry weight while ensuring hemodynamic balance. Patients were treated with intravenous antibiotic therapy com­bining two antibiotics in four cases and three antibiotics in one, administered after HD ses­sions for an average of eight weeks. Monitoring of the drug levels was performed for patients receiving vancomycin (therapeutic range between 20 and 30 g/mL). Antibiotic alone was insuffi­cient, requiring further surgical treatment by double mitral and aortic prosthetic replacement after six weeks of beginning treatment in two patients and the replacement of the aortic valve in one patient. Three of the five patients died after a mean of 10.3 months following the diag­nosis of IE.


While Sir William Osler is credited for his initial clinical description of IE, Blagg and his associates [1] were the first to report this compli­cation in HD patients. According to Friedberg, [2] the most important factor in the diagnosis of endocarditis is the physician's index of suspi­cion. This is especially true in patients recei­ving regular dialysis, whose clinical manifesta­tions may be misleading.

Infective endocarditis is at least 10-18-times more frequent in HD patients than in the ge­neral population, and its annual incidence is estimated, according to Hanslik, to be between five and 13/10,000, while the number of new cases of IE in the general population per year is estimated to be between 15,000 and 20,000 in the United States [3] and 1500-2000 in France. [4] Among patients on HD, IE caused by Staphy­lococcus aureus affecting the mitral valve caused by dual lumen catheter is the most common. [5],[6],[7],[8],[9]

Several risks factors are under investigation in HD patients, such as pre-existing damage of the endocardial valve. In 1987, Maher showed, in an ultrasonographic prospective study of 85 HD patients, that 28% of the subjects had aortic cal­cifications, 36% had mitral annular calcification and 15% had mitral and aortic calcifications, [5],[6] corresponding to what we found in our patients. The three factors associated with the occurrence of valvular calcification were age, duration on dialysis and high phosphorus-calcium product (above 5.4 mg 2 /L 2 ).

Various factors increase the susceptibility to infection in patients on HD. It may be the cau­sative factors such as diabetes, lupus and he­matological diseases, immunosuppressive treat­ment and malnutrition. Additionally, there is suppressed immune response (decrease of B and T cell response to antigenic or mitogen sti­muli and functional abnormalities of monocytes/ macrophages and neutrophils) in patients on HD, as shown by Goldman in 1990. [10]

Many writers had implicated an iron overload in the pathogenesis of these disorders. In 1990, Boelaert showed that the function of polymor­phonuclear cells in HD patients could be res­tored by reducing the iron load by using recom­binant erythropoietin. However, iron overload remains a frequent problem in HD patients, and a prospective study on 158 patients showed that the incidence of bacteremia was multiplied by three when the ferritin was more than 1000 mg/L. [11]

Clinical symptoms of IE are not typical, espe­cially in HD patients. Therefore, the physicians responsible for the care of these patients have a difficult task and must, indeed, maintain a high index of suspicion if endocarditis is to be de­tected. [12] In our study patients, symptoms were dominated by the association of fever and sys­tolic heart murmur, especially of mitral and aor­tic valves and signs of congestive heart failure, explaining the repeated episodes of APO. Al­though none of our patients had any neurolo­gical manifestations, the presences of central nervous system symptoms provide additional diagnostic clues. [13] Peripheral embolic phenomena are not pathognomonic of endocarditis, [14],[15] although their presence surely directs the physician's attention to this entity.

IE occurs more commonly one to two years after the initiation of dialysis. It usually follows access infection, particularly when accompanied by bacteremia. In our experience, IE occurred after 35.4 months following initiation of HD, and we re-emphasize the role of central catheters as a portal of entry in the pathogenesis of IE. The risk of infection is greater with catheters than by puncture of AVF.

The microbial epidemiology is characterized by the isolation of three germs: Staphylococcus aureus in 40%, Enterococcus in 33% and Strep­tococcus in 17% of the patients. [16] Staphyloco­ccus aureus is the most common pathogen, especially in patients with vascular access. [16] Fi­nally, in most series, the percentage of IE with negative blood cultures is about 10%. In one of our patients, four consecutive blood cultures re­mained negative, which could be due to the following reasons: prior antibiotic therapy, mic­roorganisms that require specific culture envi­ronments and microorganisms whose causative role in IE is usually confirmed by serology or polymerase chain reaction of the heart valves. Bacteria that are normally considered as non­pathogenic should not be interpreted as contami­nants of blood cultures until endocarditis is carefully excluded.

The mitral valve is the most common valve involved, followed by the aortic valve. The in­volvement of the tricuspid valve alone is not common [Table 5], [Table 6]. TEE has a higher sensitivity (81%) than TTE (50%) to highlight vegetations. [17],[18] In our study, the diagnosis was suspected primarily by TTE and then reinforced by TEE, which allowed appreciation of the seve­rity of the associated cardiac lesions for a better therapeutic approach.{Table 5}{Table 6}

The endocardial lesions have an anatomical specificity, which dictates the requirements of the antibiotic. It necessitates the use of a com­bination of two bactericidal antibiotics with a synergistic action. The antibiotic of choice is decided by the culture and sensitivity reports. It is necessary to get a good penetration of the antibiotic to the site of infection and, therefore, administration of intravenous antibiotics whose protein binding is low and whose diffusion is maximal on the vegetation is required. The antibiotics should be administered for four to six weeks. [19],[20]

Antibiotic alone is rarely curative. The use of surgery is necessary for a third of patients in the acute phase of IE. Although 20-40% of addi­tional patients need surgery later, [18] the decision is often difficult because of the serious condi­tion of the patients on HD.

According to Uday S. Nori et al, valve re­placement surgery did not impact survival in their patients. [16] The American College of Car­diology/American Heart Association task force defined indications for valvular surgery in pa­tients with native and prosthetic endocarditis. [21] However, the indications in patients with signi­ficant comorbid conditions are less clear. Re­cently, Spies et al [22] have reported a high peri­operative mortality in dialysis patients who underwent valve replacement surgery following IE. It is possible that this intervention may have been employed at an advanced stage of the di­sease, when serious complications had already set in and patients selected for surgery may have had severe disease.

The only specific preventive measures of en­docarditis in HD patients concerns the reduc­tion of the nasal carriage of Staphylococcus aureus by the use of a combination of rifam­picin and bacitracin or the use of muciporine; this protocol reduces the nasal carriage and de­creases the incidence of bacteremia by 75%. [23]

In conclusion, the incidence of IE is 10-18­times higher in chronic HD patients than in the general population. Infection of the vascular access, mainly catheters, is the principal risk factor in this population.


1Blagg CR, Hickman RO, Eschbach JW, Scribner BH. Home haemodialysis: six years. N Engl J Med 1970;283(21):1126-31.
2Dorney ER. Endocarditis, in Diseases of the heart, edited by Friedberg CK, 2nd ed; Philadelphia, W.B. Saunders Co, 1956;1168-84.
3Bayer AS, Bolger AF, Taubert KA, et al. Diagnosis and management of infective endo-carditis and its complications. Circulation 1998; 98(25):2936-48.
4Delahaye F, Goulet V, Lacassin F, et al. Characteristics of infective endocarditis in France in 1991. A 1-year survival. Eur Heart J 1995;16(3):394-401.
5Perdu J, Joly D, Lidove O. Med Therap 1998;4 (8):675-80.
6Hanslik T, Flahault A, Vaillant JN, et al. High risk of severe endocarditis in patients on chronic dialysis. Nephrol Dial Transplant 1997;12(6): 1301-2.
7Robinson DL, Fowler VG, Sexton DJ, Corey RG, Conlon PJ. Bacterial endocarditis in hemo­dialysis patients. Am J Kidney Dis 1997;30(4): 521-4.
8Fantin B. Physiopathologie de l'endocardite. Med Therap 1997;3:87-92.
9Hoen B. Antibiotherapie des endocardites infectieuses. Med Therap 1997;3:111-7.
10Goldman M, Vanherweghem JL. Bacterial infec­tions in chronic haemodialysis: Epidemiologic and pathophysiologic aspects. Adv Nephrol 1990;19:315-32.
11Boelaert JR, Daneels RF, Schurgers ML, Matthys EG, Gordts BZ, van Landuyt HW. Iron overload in haemodialysis patients increases the risk of bacteriemia: a prospective study. Nephrol Dial Transplant 1990;5(2):130-4.
12Harley LC, Mortan TO, Innis MD, Clunie GJ. Splenectomy for anaemia in patients on regular haemodialysis. Lancet 1971;2(7738):1343-5.
13Ziment I. Nervous system complications in bacterial endocarditis. Am J Med 1969;47(4): 593-607.
14Cross DF, Ellis JG. Occurrence of the Janeway lesions in mycotic anevrysm. Arch Intern Med 1966;118(6):588-91.
15Kilpatrick ZM, Greenberg PA, Sanford JP. Splinter Hemorrhages- Their clinical significance. Arch Intern Med 1965;115:730-5.
16Delgado-Rodriguez M, Medina-Cuadros M, Gomez-Ortega A, et al. Cholesterol and serum albumin levels as predictors of cross infection, death, and length of hospitalisation stay. Arch Surg 2002;137(7):805-12.
17Erbel R, Rohmann S, Drexler M, et al. Improvement of diagnostic value of echocardiography in patients with infective endocarditis by transoesphageal approach: a prospective study. Eur Heart J 1988;9(1):43-53.
18Quarles LD, Rutsky EA, Rostand SG. Staphylococcus aureus bacteremiain patients on chronic Heamodialysis. Am J Kidney Dis 1985;6(6):412-9
19Christopher TG, Korn D, Blair AD, Forrey AW, O'Neil MA, Cutler RE. Gentamycin pharmaco­kinetics during haemodialysis. Kidney Int 1974; 6:38-44.
20Matsuo H, Hayashi J, Ono K, et al. Administration of aminoglycosides to Heamodialysis patients immediately before dialysis: a new dosing modality. Antimicrob Agents Chemother 1997;41(12):2597-601.
21ACC/AHA guidelines for the managment of patients with valvular heart disease. A report of the American College of cardiology/ American Heart Association. Task Force on Practice Guidelines (Committee on Management of patients with valvular heart disease). J Am Coll Cardiol 1998;32(5):1486-588.
22Spies C, Madison JR, Schatz IJ. Infective endocartidis in patients with end-stage renal disease: clinical presentation and outcome. Arch Intern Med 2004;164(1):71-5.
23Boelaert JR, Van Landuyt HW, Godard CA, et al. Nasal mupirocin ointment decreases the inci­dence of Staphylococcus aureus bacteriemias in haemodialysis patients. Nephrol Dial Transplant 1993;8(3):235-9.