INTRODUCTION



In European countries, there has been in recent years a gradual increase in use of tunneled central venous catheters as a modality of permanent vascular access. According to the results reported by the DOPPS study,1 their prevalence and incidence rates are approximately 7% and 25% respectively. In our environment,2 the proportion of incident patients on hemodialysis with catheters increased at the end of the 90s from 49.5% to 55.3%, and while indwelling catheters were virtually nonexistent in 1997, they were implanted in 17.1% of patients in 2004. The proportion of prevalent patients with catheters, which was less than 8% in the 90s, has also doubled in 10 years. This exponential increase is due to the greater longevity of the current population in hemodialysis programs, as well as to late referral to the specialist and to a higher prevalence of metabolic cardiovascular disease, mainly occurring

as arterial hypertension and diabetes mellitus.3 Placement of tunneled catheters is a good alternative for definitive

vascular access in patients in whom a native arteriovenous fistula (AVF) cannot be performed or a vascular prosthesis cannot be implanted for anatomical reasons. Placement of tunneled catheters is technically simple, and they may be used immediately.4 Although blood flows (Qb) reached with tunneled catheters are increasingly high, the dialysis doses achieved are still lower than those achieved with native AFVs or vascular prostheses. This is probably related to this lower Qb and a higher number of complications associated to vascular dysfunctions and infectious processes.5-7



The primary objective of this study was to assess the additional time required to achieve an optimal dialysis dose when tunneled central venous catheters are used. Such additional time results from the lower blood flow achieved and the potential vascular changes, often requiring a decrease in Qb and the need for reversal of arteriovenous lines.



PATIENTS AND METHODS



A total of 48 patients (31 males and 17 females) with a mean age of 61.6 ± 14 years (range: 28-83) were analyzed. Patients over 18 years of age without residual kidney function on a stable hemodialysis program were enrolled into the study. There were no exclusion criteria, but 2 sessions where a great catheter dysfunction was seen were ruled out. Causes of chronic renal failure included nephroangiosclerosis (13 patients), diabetes mellitus,11 tubulointerstitial nephropathy,5 chronic glomerulopathy,5 polycystic kidney disease,5 systemic disease

(3 patients with SLE, multiple myeloma, and primary amyloidosis), bilateral nephrectomy, and unknown (5 patients). Table I shows the age, sex, and underlying conditions of the AVF and catheter groups. Among patients enrolled into the study, 20 had tunneled central venous catheters and 28 AVFs (26 natives AVFs and two PTFE prostheses). Tunneled catheter models were as follows: 12 Arrow®, 4 Split-Cath® (Medcomp ®), 2 Permcath® (Medcomp®), and 2 Canaud® (Quinton ®). Nine catheters were located in the right internal jugular vein, 8 in the left internal jugular vein, and 3 in the right femoral vein.



All patients enrolled into the study underwent high flux hemodialysis with a polysulphone 1.9 m2 dialyzer for 240 minutes with a dialysate flow of 500 mL/min. 4008 S (Fresenius) and Integra (Hospal) monitors, fitted with OCM (on-line clearance monitoring) or Diascan biosensors, were used in 42 and 6 patients respectively. Biosensors provide a noninvasive measurement of effective ionic dialysance (ID), which is equivalent to urea clearance.



The primary objective was to achieve a Kt of 45 L with each of the different vascular accesses. Patients with tunneled catheters received two dialysis sessions at maximum Qb, one with a normal line connection and the other with the lines reversed. Patients with AVFs underwent 3 sessions, in which Qb was changed to 300, 350 and 400 mL/min. Baseline ID, final ID, and Kt were recorded in each dialysis session.



Results are given as the arithmetic mean ± standard deviation.

A Student¿s t test for independent data was used to analyze the statistical significance of quantitative parameters

(figs. 1 and 2). A value of p < 0.05 was considered statistically significant.



RESULTS



The Qb values in the group of patients with tunneled catheters were 320 ± 42 mL/min in the normal position and 309 ± 46 mL/min in the reverse position. Baseline IDs were 181 ± 21 mL/min and 160 ± 20 mL/min with the catheter in the normal and reverse positions respectively (p < 0.01). In patients with native AVFs, baseline ID values were 178 ± 12, 190 ± 11, and 199 ± 17 mL/min with Qbs of 300, 350, and 400 mL/min respectively. In patients with tunneled catheters, final ID values were 163 ± 26 mL/min in the normal position and 141 ± 19 mL/min in the reverse position (p < 0.01). In patients with native AVFs, final ID values were 163 ± 16, 174 ± 10, and 179 ±

12 mL/min with Qbs of 300, 350, and 400 mL/min respectively. Figure 1 shows the baseline and final ID values obtained in patients with tunneled catheters and AVFs.



Mean clearance (K) was 171 ± 19 mL/min and 151 ± 12 mL/min with the tunneled catheter in the normal and reverse positions respectively. In patients with AVFs, mean K values were 168 ± 12, 179 ± 10, and 187 ± 11 mL/min with Qbs of 300, 350, and 400 mL/min respectively. The Kt results at session end were 40.9 ± 5 L with the catheter in normal position and 36.1 ± 4 L with the catheter reversed. Patients with AVFs achieved Kt values of 40.3 ± 3, 42.9 ± 2, and 45.0 ± 2.5 L with Qbs of 300, 350, and 400 mL/min. Figure 2 shows the Kt values obtained in patients with tunneled catheters and AVFs. Only patients with AVFs with a Qb of 200 mL/min reached the goal of a Kt of 45 L, with statistically significant differences being seen with all other vascular accesses (fig. 2). In patients with AVF, 12- and 28-minute longer hemodialysis times were required with Qb values of 350 mL/min and 300 mL/min respectively; the corresponding times for tunneled catheters in the normal and reverse positions were 24 and 59 minutes respectively. Table II summarizes the additional dialysis times required to achieve a Kt of 45 L by vascular access.



DISCUSSION



This study analyzed the additional time needed to achieve an optimal dialysis dose in patients with catheters using ID. An analysis of 48 patients, 20 of them with tunneled catheters, showed that additional dialysis times of 30 and 60 minutes were respectively required on average with the arteriovenous lines in the normal and reverse positions to achieve a Kt of 45 L. This is a crucial fact considering the gradual increase in the mean age of incident patients on hemodialysis, 80% of whom are older than 65 years.8 The immediate consequence is an increase in the proportion of temporal and indwelling catheters, higher than 50% at the start of hemodialysis.2 This has resulted in an increase in vascular access complications such as thrombosis and infectious processes and in lower dialysis

doses, leading to a higher morbidity and mortality and to an increase in current financial costs.4,9-11



While the Qbs achieved with tunneled catheters are increasingly high as compared to temporal catheters, different studies have concluded that they provide a lower dialysis dose.12,13 In this regard, the Canaud et al study5 reported interesting results showing a 5%-6% reduction in Kt/V in the group of patients using catheters. There are different reports in the literature showing that inadequate hemodialysis doses are achieved with blood flows lower than 300 mL/min.6



Current technological advances in hemodialysis allow for in situ, real time follow-up of the course of the hemodialysis session to monitor dialysis dosing and to improve dialysis tolerance. In this regard, different monitors now incorporate biosensors that measure noninvasively, using the conductivity probes of the machines, the effective ionic dialysance, which is equivalent to urea clearance (K), thus allowing for calculation of the dialysis dose without work overload, laboratory measurements, or additional cost.14 Systematic K measurement by the dialysis time elapsed allows for obtaining Kt, an actual measure of the dialysis dose, expressed in liters. Use of Kt has advantages, because both K and t are actual values measured by the monitor. If Kt/V is used, we should introduce V, and thus a value that is almost always erroneous and may be manipulated during the session. Sin 1999, Lowrie et al15,16 propose Kt as the marker of dialysis dose and mortality, recommending a minimum Kt of 40-45 L for females and 45-50 L for males. Chertow et al17 noted in 3,009 patient a J-shaped survival curve when they were distributed into quintiles based on PRU, whereas a descending curve was found when Kt was used, i.e. a higher Kt was associated to an increased survival. According to the results obtained in a previous study14 that analyzed with ID a total of 1,606 sessions in 51 patients, from 30% to 40% of patients did not reach an adequate dose, expressed

as Kt, for their sex or body surface area. Significantly, this recommended dose was not reached in 7 of the 11 patients with implanted catheters (64%).



Because of the limited number of patients enrolled into the study, differences by catheter location and type could not be assessed.



We conclude that dialyzed patients with central catheters need adjustments in dialysis time prescriptions. Because of the great dose variability between hemodialysis sessions when a catheter is used, use of monitors with ionic dialysance should ideally be generalized and Kt should be measured in each session to ensure adequate hemodialysis. When monitors allowing for Kt follow-up are not available, hemodialysis time would have to be increased, on average, by 30 minutes when a catheter is used in the normal position and by 60 minutes when in a reverse position.