Kybartienė, Sondra, Kuzminskis, Vytautas, Rimdeika, Rytis, Strupas, Kęstutis, Lukoševičius , Saulius, Vaičiūnienė, Rūta, Kaikaris, Vygintas, Pētersons , Aivars, Jankauskien, Augustina, Gulbinas, Antanas, and Lithuanian University of Health Sciences
Patients to be treated with chronic hemodialysis (HD) must have blood vessel connection. There are three types of blood vessel connections: native arteriovenous fistula (AVF), synthetic prosthesis and permanent central venous catheter. Each connection has its strengths and weaknesses. Primary AVF dysfunction occurs in 7.5% of the whole population of patients receiving HD, and in up to 80% of high risk patients (those with diabetes, elderly patients, and women) [Mann et al 2005]. Compared to other blood vessel connections (permanent central venous catheter or synthetic prosthesis) function of properly formed AVF is the longest, and incurs the lowest incidence of infectious and thrombotic complications. All international guidelines on HD procedures recommend native AVF as the method of choice. Formation of blood vessel junctions and treatment of their complications account for up to 20% of all hospitalizations of patients with end stage renal failure in USA [Allon et al 2002], up to 30% in Canada [Mann et al 2005], and up to 25% in Europe [Pissoni et al 2002]. It should be stated that formation of AVF and treatment of their complications are very expensive, and therefore determining the causes of the development of AVF complications and their timely diagnosis and correction are extremely important. Various methods of monitoring of AVF function are used in many European and USA HD centers; however, the most important criterion of early diagnosis of AVF complications has not been established yet [Paulson et al 2012]. Although large scale studies, such as DOPPS, as well as smallerscale (one country or one centre) studies that are in place in the USA and European countries look into peculiarities of AVF formation, incidence of complications and methods of their correction, no studies of AVF problems have been carried out in Lithuania. Aim and objectives Aim of the study: to determine factors related to the duration of AVF function and to evaluate benefit of monitoring AVF function in order to increase of longevity of AVF functioning. Objectives: 1. To evaluate benefits of AVF function monitoring for the increase of longevity of AVF function. 2. To determine relation between the method of the first HD procedure and duration of patient’s hospitalization. 3. To determine factors related with the duration of AVF function. 4. To determine significance of presence and site of central venous catheter to the duration of AVF function. 5. To evaluate relation between AVF maturation and duration of AVF function. 6. To evaluate significance of preoperative blood vessel ultrasound examination to the development of primary dysfunctions of fistula and duration of AVF function. Scientific novelty of the study Many nephrologists globally examine issues of AVF formation, function, and complications. However, studies examine mostly only one of above mentioned issues. In our study we analyzed problems starting from formation of AVF and the first HD procedure to AVF thrombosis and formation of new AVF. This is the first study in Lithuania examining problems of AVF formation, function and complications development. For the first time in Lithuania AVF function monitoring and preventive fistula stenosis treatment using percutaneous transluminal angioplasty (PTA) were implemented. MATERIAL AND METHODS Study was carried out in the Clinic of Nephrology of the Lithuanian University of Health Sciences; permission No. BE-2-20 of the Kaunas Region Ethical Committee was obtained. All patients with end stage renal disease treated with HD procedures in the Department of Detoxication of the Clinic of Nephrology of Kaunas Medical University from January 1, 2000 to December 31, 2010 were involved in to this study. Only patients over 18 years old were involved. Part of data has been collected retrospectively from medical documentation: personal health history (Form No. 025/a), hospital records (Form No. 003/a), and outpatients HD protocols. Other information concerning patients’ laboratory tests data and HD procedures parameters was collected during the whole study. The CRF were filled by data available for particular variable: Initial HD treatment of patient with end stage renal failure CRF. Patient's demographic data (gender, age, underlying disease causing endstage renal failure, concomitant diseases), method of the initial HD (scheduled beginning of HD through AVF or urgent procedure through central vein catheter), data concerning hospitalization for the beginning of HD procedures were collected using this CRF. Patient’s hospitalization due to AVF thrombosis CRF. Data collected using this CRF: duration of hospitallization, time to AVF formation, site of central vein catheter and its complications, site of AVF, its maturation time and data collected during blood vessels preoperative ultrasound examination. Arteriovenous fistula monitoring CRF. This CRF was filled in only for monitored patients. Patient’s laboratory blood tests data, HD efficacy marker, HD procedure parameters, and AVF fistula examination data were collected using this questionnaire. Principles of monitoring of arteriovenous fistula Parameters of every HD procedure were recorded in HD procedure protocols (they are filled in by HD center staff during HD procedures). Since January 1, 2009 arterial and venous blood flow resistance was recorded in these protocols; this parameter has been measured 15 minutes after the beginning of HD procedure when volume of blood passing through HD apparatus was 200 ml/min. Parameters of the procedure performed during the last week of the month were recorded in the questionnaire. Nurse working in the HD center performed clinical examination of AVF and in case of changes informed investigator on the same day. Indications for fistula ultrasound examination At least one of the following: 1. Kt/V decrease 0.4–0.5 when there are no other objective reasons for decrease. 2. 20–30% decrease of volume of blood flowing through HD apparatus during procedure when there are no other objective reasons for decrease. 3. >50% increase of arterial and venous blood flow resistance during one month when there are no other objective reasons for increase. 4. All abnormal findings during clinical examination of fistula. During our study all ultrasound examinations were performed by one investigator, radiologist, who had several years experience of AVF examination, the same ultrasound machine was used for ultrasound examinations. Indications for percutaneous transluminal angioplasty of the fistula At least one of the following: 1. Diameter of AVF anastomosis < 2.0 mm; 2. Peripheral stenosis of venous part of AVF with or without partial thrombosis; 3. Partial AVF thrombosis; 4. Volume of blood flowing through brachial artery 25% decrease of blood flow through brachial artery through time. PTA procedures were performed within one week from the establishing of indications for this procedure. All procedures were performed by one physician, invasive radiologist, who had several years experience in the treatment of AVF complications. Statistical data analysis Statistical analysis of study data has been performed using IBM SPSS statistical software package 19.0 version. To test a hypothesis that quantitative values of population are distributed according to Gauss distribution we used modified Kolmogorov-Smirnov test with Lillefors correction. To describe normally distributed (Gauss distribution) quantitative values we used main characteristic of position, mean with standard error (±SE), and standard deviation (s) characterizing dispersion of data. To describe normally distributed rank and qualitative values we calculated median value. Nominal values were described using relative frequencies. To compare means of several normally distributed (Gauss distribution) quantitative values we used parametric criteria (t-test, ANOVA), and in order to test if attributes of not normally distributed rank and qualitative values are uniform we use nonparametric criteria (Mann-Whittney, Cruscal-Wollis rank criteria). Confidence level was P=0.95, and significance level was α=p=0.05. To evaluate statistically significant dependence of two quantitative values and their homogeneity of distribution within comparison group we used χ2 test for independence (homogeneity). Strength of dependence of two quantitative values within the sample was evaluated by calculating Pearson contingency coefficient. Probability of qualitative value within two analyzed groups was compared using criteria of parity of probabilities (Z-test). Time period from the formation of AVF to its thrombosis or development of critical stenosis when it was necessary to form a new fistula we called AVF survival. In order to present diagram of AVF survival we used Kaplan-Meier method. Log-rank criterion (log-rank test) was used to compare these functions. Influence of single attributes (factors influencing AVF survival) on the fistula life time was evaluated during analysis of AVF survival. Model of proportional risks (Cox model) was used for this purpose. CONCLUSIONS 1. Monitoring of AVF function statistically significantly improved one year survival of fistulas. Thrombosis risk was 0.245 times lower in monitoring group fistulas compared to historical group; thrombosis risk in partially monitored group was 0.132 times lower compared with historical group. 2. Duration of hospital stay for patients who underwent urgent HD treatment through central venous catheter was statistically significantly longer (average duration of hospital stay was 33.15±1.55 days) compared to patients who underwent HD through matured AVF (average duration of hospital stay was 15.12±1.15 days) (p