Septicemia in dialysis patients: incidence, risk factors, and prognosis

Dialysis access related infections

Infection is a common cause of morbidity and mortality in end-stage renal disease patients. Unintentional pathogens are introduced into an immunocompromised host during hemodialysis and peritoneal dialysis by means of the access (arteriovenous fistula, arteriovenous graft, central venous catheter, or peritoneal dialysis catheter). Gram positive organisms are most common with Staphylococcus aureus and coagulase negative Staphylococcus predominating. Preventive measures are mandatory and the key to decreasing infection rates.

Dialysis access infections

Infections and specifically infectious complications of vascular access remain a major cause of morbidity and mortality in the hemodialysis population. Primary arteriovenous fistulas have the lowest rates of infections and are the access of choice whenever vascular anatomy allows. The dialysis outcomes quality initiative (DOQI) guidelines have thus stressed the need for increasing the utilization of arteriovenous fistulas. Unfortunately, comorbid disease processes and late referrals for vascular access have maintained our dependence on synthetic grafts and indwelling catheters. Indwelling catheters, in particular, have the highest rate of infection and are often associated with more serious metastatic complications. Appropriate antibiotics along with aggressive surgical debridement remain crucial in bacteremia occurring in arteriovenous fistulas or synthetic grafts (polytetrafluoroethylene). Catheter related bacteremia necessitates catheter removal with either guidewire exchange or replacement after a period of antibiotic therapy. Measures to increase our utilization of primary fistulas whenever possible will lower the risk of these complications in our patients.

Infective endocarditis in patients receiving long-term hemodialysis

OBJECTIVE

To ascertain the predominant characteristics of patients receiving long-term dialysis who develop infective endocarditis (IE).

PATIENTS AND METHODS

We reviewed the records of all chronic hemodialysis patients who had IE at Mayo Clinic, Rochester, Minn, between 1983 and 1997.

RESULTS

Twenty episodes of IE occurred in 17 patients. One patient had 3 episodes of IE, and 1 patient had 2 episodes of IE; each episode was caused by a different organism. The mean +/- SD age of our patients was 63 +/- 11 years; there were 13 males; 6 patients had diabetes mellitus; and the mean +/- SD duration of hemodialysis prior to IE was 24.2 +/- 20.5 months. This analysis included 10 episodes of IE (occurring in 9 patients) within the Mayo Clinic Dialysis System during which time 223,358 hemodialysis treatments were delivered, giving a rate of 10 IE episode per 223,336 hemodialysis treatments. Among all 20 IE episodes, there were 14 synthetic arteriovenous grafts, 4 permanent venous dialysis catheters, 2 temporary venous dialysis catheters, and 2 native arteriovenous fistulas (2 accesses in 2 patients), and access had been in place for a mean +/- SD of 15.9 +/- 18.6 months. The portal of infection was the hemodialysis access in 13 episodes of IE. The causative organisms for IE were Staphylococcus aureus in 8 cases, Enterococcus sp in 4 cases, viridans streptococcus in 3 cases, Staphylococcus epidermidis in 2 cases, and 1 case each of Streptococcus bovis, group G beta-hemolytic streptococcus, and Aspergillus sp. The mitral valve was involved in 9 cases, the aortic valve was involved in 5 cases, and the tricuspid and pulmonic valves were involved in 1 case each. Patient survival (after the first episode of IE) was 71% at 30 days; 53% at 60 days; and 35% at 1 year. Echocardiography was performed in 19 episodes of IE. The transthoracic echocardiogram was 62.5% sensitive and 40% specific for the presence of definite or probable vegetations. Univariate analysis for factors affecting 60-day survival show that presence of right-sided IE, vegetation size greater than 2.0 cm3, diagnosis of diabetes mellitus, and initial leukocyte count greater than 12.5 x 10(9)/L were poor prognostic factors. Aortic valve involvement carried a better prognosis.

CONCLUSIONS

Infective endocarditis in hemodialysis patients is relatively infrequent but has a high mortality. Patients with synthetic intravascular dialysis angioaccess (synthetic grafts and venous catheters) are more likely to develop IE than patients with native arteriovenous fistulas. Transesophageal echocardiography is a preferred echocardiographic study for suspected cases of IE. Prolonged antibiotic therapy is needed for all patients, and close monitoring is needed for patients with right-sided IE, large vegetations, diabetes mellitus, and an elevated leukocyte count.

Mortality caused by sepsis in patients with end-stage renal disease compared with the general population

BACKGROUND

In the United States, infection is second to cardiovascular disease as the leading cause of death in patients with end-stage renal disease (ESRD), and septicemia accounts for more than 75% of this category. This increased susceptibility to infections is partly due to uremia, old age, and comorbid conditions. Although it is intuitive to believe that mortality caused by sepsis may be higher in patients with ESRD compared with the general population (GP), no such data are currently available.

METHODS

We compared annual mortality rates caused by sepsis in patients with ESRD (U.S. Health Care Financing Administration 2746 death notification form) with those in the GP (death certificate). Data were abstracted from the U.S. Renal Data System (1994 through 1996 Special Data request) and the National Center for Health Statistics. Data were stratified by age, gender, race, and diabetes mellitus (DM). Sensitivity analyses were performed to account for potential limitations of the data sources.

RESULTS

Overall, the annual percentage mortality secondary to sepsis was approximately 100- to 300-fold higher in dialysis patients and 20-fold higher in renal transplant recipients (RTRs) compared with the GP. Mortality caused by sepsis was higher among diabetic patients across all populations. After stratification for age, differences between groups decreased but retained their magnitude. These findings remained robust despite a wide range of sensitivity analyses. Indeed, mortality secondary to sepsis remained approximately 50-fold higher in dialysis patients compared with the GP, using multiple cause-of-death analyses; was approximately 50-fold higher in diabetic patients with ESRD compared with diabetic patients in the GP, when accounting for underreporting of DM on death certificates in the GP; and was approximately 30-fold higher in RTRs compared with the GP, when accounting for the incomplete ascertainment of cause of death among RTRs. Furthermore, despite assignment of primary cause-of-death to major organ infections in the GP, annual mortality secondary to sepsis remained 30- to 45-fold higher in the dialysis population.

CONCLUSIONS

Patients with ESRD treated by dialysis have higher annual mortality rates caused by sepsis compared with the GP, even after stratification for age, race, and DM. Consequently, this patient population should be considered at high-risk for the development of lethal sepsis.

A pseudo-outbreak of Aureobasidium species lower respiratory tract infections caused by reuse of single-use stopcocks during bronchoscopy

OBJECTIVE

To investigate and control an apparent outbreak of lower respiratory tract infections due to Aureobasidium species.

DESIGN

Outbreak investigation.

SETTING

University-affiliated medical center.

PATIENTS

Nine patients who underwent bronchoscopy between June and August 1998.

RESULTS

Ten bronchoalveolar lavage (BAL) fluid cultures from nine patients grew Aureobasidium species during the outbreak period; whereas, respiratory specimens from only two patients grew Aureobasidium species during the preceding 6 years. No patient was judged to have true infection due to Aureobasidium species either before or after bronchoscopy. Nine of the 10 bronchoscopies that yielded Aureobasidium species were performed in the outpatient bronchoscopy suite. The Aureobasidium isolates were not associated with any one bronchoscope. Observation of bronchoscopy procedure revealed that plastic stopcocks labeled for single use were reused on different patients during BAL. There was no record of how many times each stopcock was being reused. After each use, the stopcocks were placed in an automated disinfection machine designed for bronchoscopes. Culture of the stopcocks after they had been "disinfected" yielded a heavy growth of Aureobasidium species, while culture of fluid from the automated disinfection machine was negative. Reuse of the stopcocks was halted, and, during the following 6-month period, Aureobasidium species were not isolated from any BAL specimen.

CONCLUSIONS

Reuse of medical equipment labeled for single use is potentially hazardous, especially if no quality control system is in place to monitor sterility and function after reprocessing.

Septicemia in diabetic hemodialysis patients: comparison of incidence, risk factors, and mortality with nondiabetic hemodialysis patients

Diabetes mellitus is the most common cause of treated end-stage renal disease (ESRD), and diabetic hemodialysis patients have a high mortality rate. To identify differences in risk of septicemia among diabetic and nondiabetic hemodialysis patients, we examined the incidence, risk factors, and mortality for septicemia in a large sample of the US hemodialysis population. We performed a longitudinal cohort study of the incidence and risk factors for hospitalized cases of septicemia in diabetic and nondiabetic hemodialysis patients using baseline data from the US Renal Data System case-mix severity study with 7-year follow-up from hospitalization and death records. Independent risk factors for septicemia were assessed using Poisson regression. Independent effect of septicemia on mortality was assessed using Cox proportional hazards analysis. Over 7 years, 11.1% of nondiabetic patients and 12.5% of diabetic patients experienced at least one episode of septicemia. Older age and low serum albumin were independent risk factors for septicemia in all patients. In diabetics, white race, peripheral vascular disease, and hemodialyzer reuse, particularly in type 1, were independent risk factors. In nondiabetics, coronary artery disease, cerebrovascular disease, and temporary and permanent catheters were associated with an increased risk. In both groups, patients who experienced an episode of septicemia had twice the risk of death from any cause and an eightfold risk of death from septicemia. Septicemia occurs equally frequently and carries a marked increased risk of death in both nondiabetic and diabetic hemodialysis patients. Improving nutritional status and minimizing the use of catheters might help ameliorate the risk of septicemia. In diabetics, aggressive treatment of peripheral vascular disease might help reduce the risk of septicemia. Further research to elucidate potential mechanisms for variations in risk for septicemia according to race and hemodialyzer reuse practices are warranted in diabetic patients.

Haemodialysis access: the case for prospective monitoring

Thrombosis in haemodialysis accesses remains a major problem. It is associated with stenosis that causes haemodynamic and anatomical changes. By prospective monitoring it is possible to identify patients at risk of thrombosis. Those patients should be referred for corrective intervention. This approach can result in a thrombosis rate below the advised quality of care standard of 0.5 thromboses/patient-year.