Volume 20, Issue 12 , Pages 1578-1581, December 2009
The Case for Primary Placement of Tunneled Hemodialysis Catheters in Acute Kidney Injury
Article Outline
Purpose
Nontunneled hemodialysis catheters (NTDCs) are widely used for initial hemodialysis access in new-onset renal failure. The National Kidney Foundation recommends NTDC use for hemodialysis duration of less than 1 week in acute kidney injury because of the increased infection risk compared with tunneled hemodialysis catheters (TDCs) with longer use. The present study was performed to determine whether primary placement of TDCs in this setting is more appropriate, and whether there are predictors of recovery of renal function in less than 1 week.
Materials and Methods
In the authors' practice, patients referred to the interventional radiology unit in whom no contraindications exist receive a TDC; 76 patients who received a primary TDC for acute kidney injury and who eventually recovered renal function were retrospectively reviewed herein. Causes of renal failure, various renal function parameters, and demographics were collected, as were TDC dwell times, in an effort to determine predictors of recovery and/or extended duration of use.
Results
Mean TDC dwell time in patients who eventually recovered from acute kidney injury was 34 days; only 15 of 76 (20%) recovered within 1 week. At TDC placement, there were no significant differences between patients who recovered in less than (vs greater than) 1 week.
Conclusions
The present results support primary placement of TDCs in patients with acute kidney injury who require hemodialysis and in whom no contraindications exist, as no predictors of recovery of renal function in less than 1 week were identified.
Abbreviations: NTDC, nontunneled hemodialysis catheter, TDC, tunneled hemodialysis catheter
NONTUNNELED hemodialysis catheters (NTDCs) are widely used as initial hemodialysis access in new-onset renal failure (1, 2). The most recent National Kidney Foundation Kidney Disease Outcomes Quality Initiative recommendations for hemodialysis access in acute kidney injury (the preferred term rather than “acute renal failure”) state that tunneled hemodialysis catheters (TDCs) should be used for hemodialysis durations greater than 1 week and that conversion of NTDCs to TDCs should be performed if the duration of hemodialysis exceeds 1 week. It is not clear whether the current guidelines express a preference for NTDCs versus TDCs in the 1-week range. The recommendation to limit NTDC use to short duration stems from increased risk of infection compared with TDCs, as well as greater blood flow rates achievable with TDCs (3, 4, 5, 6, 7, 8, 9, 10, 11). Beyond 1 week, the risk of NTDC infection increases substantially. Patency and infection-free survival rates within 2 weeks of TDC placement are superior to those associated with NTDCs, with an infection rate per 1,000 catheter-days more than five times greater for internal jugular NTDCs than TDCs (12).
Accordingly, the decision for primary placement of an NTDC or a TDC in acute kidney injury hinges on whether renal function is expected to return in less than 1 week or greater than 1 week. However, few studies have examined potential predictors of renal recovery in patients with acute kidney injury (13, 14, 15, 16) and, consequently, estimating the temporal course of renal recovery remains difficult.
Given the uncertainty concerning the duration of renal recovery in this population, we sought to determine whether primary placement of TDCs in this setting is more appropriate than NTDC placement, and whether we could identify any predictors of recovery of renal function in less than 1 week.
Materials and Methods
We conducted a retrospective study to assess the relationship between potential demographic and clinical predictors of renal recovery and duration of renal recovery in patients with acute kidney injury requiring hemodialysis. This study received institutional review board approval and met waiver requirements for the Health Insurance Portability and Accountability Act.
In our practice, by mutual consent of the hemodialysis access team, all patients referred to the interventional radiology unit in whom no contraindications to TDC placement exist (eg, uncorrectable coagulopathy with International Normalized Ratio >2, thrombocytopenia with platelet count <25,000, bacteremia) receive a TDC. Through a quality assurance database from October 2001 to April 2007, 111 patients were identified who received a primary TDC (ie, the TDC was the first catheter placed) for acute kidney injury and who eventually recovered renal function. Specifically, the principal inclusion criterion was removal of a TDC for recovery of renal function. Patients were subsequently excluded from this study if they (i) had a TDC previously placed at an outside hospital (n = 4), (ii) had been prematurely identified as having recovered and subsequently required another TDC within the same hospitalization (n = 9), (iii) had a TDC prematurely removed or replaced for fever/infection (n = 6) or poor flow (n = 3), (iv) had medical charts containing incomplete data (n = 10), or (v) had medical charts that could not be obtained (n = 3). Given these exclusion criteria, the final study population consisted of 76 patients.
Patient charts were reviewed to determine when the decision was made by a renal attending physician to place the TDC. Data for potential predictors of renal recovery were then abstracted from patient charts at this decision point, including serum creatinine level, 24-hour urine output, oliguria duration if applicable (defined as <400 mL in 24 hours), etiology of renal failure, active comorbidities, age, sex, and patient location within the hospital. The dependent variable of interest, TDC dwell time—a proxy for duration of renal recovery—was calculated according to the date the TDC was placed and the date the TDC was removed in the interventional radiology unit. A renal attending physician made the decision when renal function had recovered and therefore when the TDC for hemodialysis should be removed.
All catheters were 14.5-F split-tip catheters (Ash; Medcomp, Harleysville, Pennsylvania) and all were placed by fellowship-trained, Certificate of Added Qualification–certified or -eligible interventional radiologists or by residents/fellows with direct attending physician supervision. Catheters were removed with use of the traction technique as described elsewhere (17).
Statistical Analysis
Associations between potential predictors of interest and TDC dwell time were assessed with use of the Student t test and analysis of variance for continuous variables and the χ2 test for categoric variables. Statistical analyses were conducted with use of Stata software (version 9.1; Stata, College Station, Texas).
In the nine patients excluded because the TDC was prematurely replaced for fever/infection or poor flow, we performed a separate analysis including the six who had over-the-wire exchanges (ie, we continued to exclude three of those patients who had the TDC removed and subsequently placed again at a later date). The rationale for doing so was that we were unsure how best to treat these patients who technically had the original TDC removed, yet immediately received a replacement that remained until recovery of renal function.
Results
There were no insertion complications in the study. Mean TDC dwell time in patients who eventually recovered from acute kidney injury was 34 days; only 15 of 76 (20%) recovered within 1 week. Thirty-three of 76 patients (43%) recovered within 2 weeks, 45 (59%) within 3 weeks, and 56 (74%) within 4 weeks. The remaining 20 patients (26%) recovered in longer than 4 weeks. When including the six patients who had over-the-wire exchanges for fever/infection (n = 3) or poor flow (n = 3), the mean TDC dwell time increased to 35 days, and only 15 of 82 patients (18%) showed recovery within 1 week. At the time of TDC placement, mean serum creatinine level was 6.2 mg/dL, 24-hour urine volume was 781 mL, and duration of oliguria was 2 days among the 22 patients (29%) with oliguria. Patients who recovered in less than 1 week (n = 15) versus greater than 1 week (n = 61) were similar (P = .37) in terms of hypothesized predictors of renal recovery (Table 1).
Table 1. Characteristics of Patients by Duration of Renal Recovery
| Characteristic | Recovery <1 Week (n = 15) | Recovery ≥1 Week (n = 61) | P Value |
|---|---|---|---|
| Age (y) | 56 ±4.9 | 56±1.9 | .99 |
| Female sex (%) | 40 | 39 | .96 |
| Creatinine (mg/dL) | 6±0.54 | 6±0.4 | .7 |
| 24-hour urine volume (mL) | 751±201 | 791±118 | .87 |
| Oliguria duration (d) | 2.4±1.4⁎ | 1.9±1.3† | .37 |
| Chronic renal insufficiency | 8/15(53) | 22/52(42) | .45 |
| Location in hospital (%) | .14 | ||
| Medicine floor | 10(67) | 24(39) | |
| Cardiac care unit | 3(20) | 5(8) | |
| Medical intensive care unit | 1(7) | 12(20) | |
| Intermediate medical care unit | 1(7) | 2(3) | |
| Cardiac intensive care unit | 0 | 4(7) | |
| Surgical intensive care unit | 0 | 4(7) | |
| Unknown | 0 | 10(16) |
⁎Among seven of 15 patients (47%) with oliguria. |
†Among 15 of 61 patients (25%) with oliguria. |
The distribution of etiologies of renal failure in patients who recovered from acute renal injury in less than 1 week was not significantly different from that in patients whose recovery took longer than 1 week (Table 2). Of note, eight of 15 patients (53.3%) who recovered in less than 1 week had underlying chronic kidney disease per their patient chart, similar to the 22 of 52 patients (42.3%) who recovered in greater than 1 week (P = .45).
Table 2. Etiology of Renal Failure by Duration of Renal Recovery⁎
| Etiology | <1 Week (n = 15) | ≥1 Week (n = 61) |
|---|---|---|
| Hypovolemia/hypotension | 3 | 6 |
| Drug toxicity | 2 | 2 |
| Heart failure | 2 | 2 |
| Hepatorenal syndrome | 1 | 2 |
| Hypovolemia/hypotension, drug toxicity | 1 | 2 |
| Contrast nephropathy | 1 | 1 |
| Contrast nephropathy, infection, hypovolemia/hypotension | 1 | 0 |
| Infection, hypovolemia/hypotension | 1 | 0 |
| Sepsis, heart failure | 1 | 0 |
| Shock, heart failure, drug toxicity | 1 | 0 |
| Tumor lysis, shock | 1 | 0 |
| Rhabdomyolysis | 0 | 5 |
| Acute kidney injury on chronic renal insufficiency | 0 | 3 |
| Acute kidney injury on chronic renal insufficiency, hypovolemia/hypotension | 0 | 2 |
| Heart failure, hypovolemia/hypotension | 0 | 2 |
| Membranoproliferative glomerulonephritis | 0 | 2 |
| Multiple myeloma | 0 | 2 |
| Rapidly progressive glomerulonephritis | 0 | 2 |
| Thrombotic thrombocytopenic purpura | 0 | 2 |
| Contrast nephropathy, heart failure | 0 | 2 |
| Heart failure, acute kidney injury on chronic renal insufficiency | 0 | 1 |
| Heart failure, hypovolemia/hypotension, contrast nephropathy | 0 | 1 |
| Heart failure, rapidly progressive glomerulonephritis | 0 | 1 |
| Hepatorenal syndrome, drug toxicity | 0 | 1 |
| Hepatorenal syndrome, hypotension/hypovolemia | 0 | 1 |
| HIV nephropathy | 0 | 1 |
| Multiple myeloma, membranoproliferative glomerulonephritis, hepatorenal syndrome | 0 | 1 |
| After renal obstruction | 0 | 1 |
| Sepsis | 0 | 1 |
| Sepsis, drug toxicity | 0 | 1 |
| Sepsis, hypovolemia/hypotension | 0 | 1 |
| Systemic lupus erythematosus nephritis | 0 | 1 |
| Tumor lysis | 0 | 1 |
| Unknown | 0 | 10 |
|
⁎Analysis of variance yields a global P value of .48, which is not significant, so there are no associations for any etiologies. Although no association between etiology and recovery time was found, this should be interpreted with caution because the study is not adequately powered. |
It was not the purpose of this study to determine outcomes of TDCs overall; this is well established. Indeed, by definition, infected catheters were not included in the series, as described earlier. Nonetheless, the rate of infection requiring removal in this series can be approximated based on the 2,870 catheter days for the 82 catheters and three infections in this group at one per 1,000 catheter-days.
Alternately, based on the 111 patients and a mean dwell time of 34 days (ie, 3,774 catheter-days), a rate of 1.6 per 1,000 catheter days can be estimated. Therefore, with a rate between 1 and 1.6 infections per 1,000 catheter-days, the infection rate in the present study is well within the range in published reports.
Discussion
Widespread use of NTDCs as initial hemodialysis accesses in cases of new-onset renal failure was initially motivated by infection rates comparable to those associated with TDCs for 1–3 weeks after placement, low invasiveness, and cost (1, 2). Recently, the National Kidney Foundation updated their Kidney Disease Outcomes Quality Initiative recommendations to recommend that NTDCs should not remain in place for more than 1 week as a result of the increased risk of infection with prolonged NTDC dwell time (3, 12). The mean TDC dwell time in our patients who recovered renal function was 4 weeks longer than the Kidney Disease Outcomes Quality Initiative maximum recommended NTDC dwell time of 1 week, and 80% of patients did not show recovery of renal function by 1 week. If those 80% of patients had been given a NTDC at the decision point, replacement of the NTDC with an TDC would have been required after 1 week. By placing TDCs at the decision point instead, these 80% of patients potentially avoided risks, morbidities, and costs associated with a second procedure. Other investigators have reported hemodialysis durations greater than 1 week as well. The mean duration of intensive renal replacement therapy in critically ill patients was 13.4 days ± 9.6 days—and 12.8 d ± 9.3 with less intensive therapy—in the study of Palevsky et al (18). The mean hemodialysis duration in critically ill patients was 11 days for continuous and intermittent hemodialysis in the study of Vinsonneau et al (19).
Several factors have previously been suggested that may shorten the duration of renal recovery, including younger age, male sex, low burden of comorbid illness, normal premorbid renal function, severe sepsis/septic shock, and lower therapy serum creatinine levels before renal replacement (13). One study (15) found age greater than 65 years to be a predictor of impaired recovery of renal function after acute kidney injury. In contrast, a prospective study (16) showed that neither patient characteristics (eg, age, sex, comorbid conditions), severity of illness (eg, Acute Physiology and Chronic Health Evaluation III status, number of failed organs), nor mode and duration of renal replacement therapy were related to recovery of renal function. Similarly, we were unable to identify any predictors of renal recovery in the present study, possibly as a result of insufficient power to detect statistically significant differences between patients whose renal function recovered in less than 1 week versus longer than 1 week. We believe the benefit to the 80% of patients who needed TDCs in our study (ie, avoiding multiple procedures and catheter changes and thereby increased risk of infection and thrombosis) outweighs the increased cost to the 20% who could have received NTDCs in retrospect.
Other important limitations to the present study deserve mention. As a result of the retrospective nature of the study, we cannot be certain how much (if any) lag time existed between when renal function actually recovered and when the TDCs were removed. In addition, because it was not a focus of the study, the actual number of hemodialysis sessions provided by each catheter was not specifically studied. Because of the complications associated with hemodialysis catheters, we are very aggressive about removing them when they are no longer needed; therefore, we believe this limitation is minimal. Another limitation is that patients included in the study had many active comorbidities, which limited our ability to determine whether a low burden of comorbid illness is in fact a predictor of renal function recovery within 1 week. In addition, our sample set was limited to those who had primary TDCs placed and whose indication for removal was recovery of renal function. We therefore have an inherent selection bias, as we intentionally did not include any patients with contraindications to a TDC, including uncorrectable coagulopathy, thrombocytopenia, and bacteremia. Such patients would have received a NTDC at our institution. Therefore, any conclusions drawn from this study cannot be extended to such patients. In addition, our study included only patients referred to the interventional radiology unit for placement of a hemodialysis catheter even though NTDCs are occasionally placed at the bedside in intensive care units when the access is needed emergently or the patient is not in stable enough condition to be transported to the interventional radiology unit. Our service places the vast majority of NTDCS and all TDCS in our hospital, so we believe the few patients undergoing bedside placement—although they were not included in the study—likely would not have affected the outcome of the study. We recognize and acknowledge all these limitations and suggest that a prospective randomized study comparing NTDCs and TDCs in patients with acute kidney injury would be the ideal way to evaluate the risk/benefit ratios of each approach.
Our results argue for primary placement of TDCs in all patients with acute kidney injury who require hemodialysis and in whom no contraindications exist, as we were unable to identify any predictors of recovery of renal function in less than 1 week. This suggests that revision of current Kidney Disease Outcomes Quality Initiative recommendations may be warranted.
References
- . Risk of bacteremia from temporary hemodialysis catheters by site of insertion and duration of use: a prospective study. Kidney Int. 2000;58:2543–2545
- . Central venous catheters: many questions, few answers. Nephrol Dial Transplant. 2002;17:1368–1373
- . NKF K/DOQI Guidelines (2006 Updates) – Vascular Access. http://www.kidney.org/professionals/KDOQI/guideline_upHD_PD_VA/va_guide2.htmAccessed January 19, 2009
- . Role of vascular access as a risk factor for infections in hemodialysis. Ren Fail. 2003;25:967–973
- . Central venous angioaccess for hemodialysis and its complications. Semin Dial. 1992;5:121–128
- . Adequacy of haemodialysis with cuffed central-vein catheters. Nephrol Dial Transplant. 1998;13:745–749
- . Dialysis access infections. Curr Opin Nephrol Hypertens. 2000;9:631–635
- . Subclavian hemodialysis catheter infections. Am J Kidney Dis. 1986;7:421–427
- . Factors affecting delivery of high-efficiency dialysis using temporary vascular access. Am J Kidney Dis. 1993;22:24–29
- . The use of dual lumen jugular venous catheters as definitive long term access for haemodialysis. Int J Artif Organs. 1990;13:26–31
- . New national surveillance system for hemodialysis-associated infections: initial results. Am J Infect Control. 2002;30:288–295
- . Compared to tunnelled cuffed haemodialysis catheters, temporary untunnelled catheters are associated with more complications already within 2 weeks of use. Nephrol Dial Transplant. 2004;19:670–677
- . Epidemiology of renal recovery after acute renal failure. Curr Opin Crit Care. 2006;12:544–550
- . Renal recovery after severe acute renal failure. Int J Artif Organs. 2006;29:1023–1030
- . Recovery of kidney function after acute kidney injury in the elderly: a systematic review and meta-analysis. Am J Kidney Dis. 2008;52:262–271
- . Renal recovery from acute tubular necrosis requiring renal replacement therapy: a prospective study in critically ill patients. Nephrol Dial Transplant. 2006;21:1248–1252
- Outcome of polyester cuff retention following traction removal of tunneled central venous catheters. Radiology. 2001;219:651–654
- VA/NIH Acute Renal Failure Trial Network (Intensity of renal support in critically ill patients with acute kidney injury). N Engl J Med. 2008;359:7–20
- Continuous venovenous haemodiafiltration versus intermittent haemodialysis for acute renal failure in patients with multiple-organ dysfunction syndrome: a multicentre randomised trial. Lancet. 2006;368:379–385
From the SIR 2008 Annual Meeting.
S.O.T. is a paid consultant for MedComp (Harleysville, Pennsylvania), Arrow/Teleflex (Reading, Pennsylvania), and Bard (Tempe, Arizona). None of the other authors have identified a conflict of interest.
PII: S1051-0443(09)00875-6
doi:10.1016/j.jvir.2009.08.014
© 2009 SIR. Published by Elsevier Inc. All rights reserved.
Volume 20, Issue 12 , Pages 1578-1581, December 2009