Different Prescribed Doses
of High-Volume Peritoneal
Dialysis and Outcome of
Patients with Acute Kidney
Injury
The optimal dialysis dose for the treatment of acute kidney injury (AKI) is controversial. No studies have directly examined the effects of peritoneal dialysis (PD) dose on outcomes in AKI. From January 2005 to January 2007, we randomly as
-signed critically ill patients with AKI to receive higher- or lower-intensity PD therapy (prescribed Kt/V of 0.8 and 0.5 per session respectively). The main outcome measure was death within 30 days. Of the 61 enrolled patients, 30 were randomly assigned to higher-intensity therapy, and 31, to a lower-intensity PD dose. The two study groups had similar baseline characteristics and received treatment for 6.1 days and 5.7 days respectively
(p = 0.42). At 30 days after randomization, 17 deaths had occurred in the higher-intensity group (55%), and 16 deaths, in the lower-intensity group (53%, p = 0.83). There was a significant differ
-ence between the groups in the PD dose prescribed compared with the dose delivered (higher-intensity group: 0.8 vs. 0.59, p = 0.04; lower-intensity group: 0.5 vs. 0.49, p = 0.89). The groups had similar metabolic control after 4 PD sessions (blood urea nitrogen: 69.3 ± 14.4 mg/dL and 60.3 ± 11.1 mg/dL respectively, p = 0.71). In critically ill patients with AKI, an intensive PD dose did not lower the mortality or improve the recovery of kidney function or metabolic control. The PD dose is limited by dialysate flow and membrane perme
-ability, and clearance per exchange can decrease if a shorter dwell time is applied.
Key words
Acute kidney injury, dialysis volume, mortality
Introduction
The optimal approach to renal replacement therapy (RRT) and the optimal intensity and timing of such therapy in critically ill patients remain unclear. The use of peritoneal dialysis (PD) is widespread in develop-ing countries in view of its ease of use, low cost, and minimal infrastructure requirements (1–3).
A recent trial by Gabriel et al. (4) showed that high-volume PD is comparable to daily hemodialysis (HD) in the treatment of patients with acute kidney injury (AKI), and those authors suggested that the use of PD was associated with more rapid renal recovery. But it is important to irst determine what the dose targets are.
In a single-center randomized controlled study in which continuous RRT was the sole treatment approach, survival improved when the intensity of therapy was increased from an assigned hourly efluent rate of 25 mL per kilogram of body weight to either 35 mL or 45 mL (5). However, subsequent studies have produced conlicting results (6–8). The recently reported Veterans Affairs and National Institutes of Health Acute Renal Failure Trial Network study (9) showed that increasing the intensity of RRT did not lower the mortality among patients with AKI. Further-more, few studies have directly examined variation of the dose of PD in AKI. There are concerns that PD is unable to achieve the high clearances required to support a patient with renal failure. However, various techniques that are able to achieve the necessary small-solute clearance targets have been described. Those techniques include high-volume PD and continuous-low PD. The selection of lexible peritoneal catheters with better catheter function and dialysate low rates can improve the eficiency of PD (4,10).
We conducted a randomized controlled study to test the hypothesis that increasing the intensity of Daniela Ponce, Germana Alves Brito, Juliana Gera Abrão,
André Luis Balbi
Ponce et al. 119
the PD dose would reduce mortality at 30 days and improve metabolic control.
Methods
Study design
Our prospective randomized trial was designed to assess two levels of intensity in high-volume PD therapy for critically ill patients with AKI. The study was conducted between January 2005 and January 2007 at the Botucatu School of Medicine, Sao Paulo State University, Brazil.
The study population comprised adults with AKI according to Acute Kidney Injury Network criteria (11). The main inclusion criterion was a clinical diagnosis of septic AKI and severe acute tubular necrosis caused by a recent ischemic or nephrotoxic injury. Patients were excluded if they were less than 18 years of age; if they had functional azotemia, urinary tract obstruction, acute interstitial nephritis, rapidly progressive glomerulonephritis, a history of chronic renal insuficiency (serum creatinine greater than 4 mg/100 mL), renal transplantation, pregnancy, severe hypercatabolism according to Schrier criteria (11), or an absolute contraindication for PD [recent abdominal surgery (<1 month) or multiple abdomi-nal surgeries (>3)]; and if they had undergone less than 1 session (deined as 24 hours) of high-volume PD (HVPD).
The indications for dialysis were uremic symp-toms, a blood urea nitrogen (BUN) level greater than 100 mg/dL (azotemia), volume overload, electrolyte imbalance (potassium > 6 mEq/L after clinical treat-ment), or refractory acid–base disturbance (bicarbon-ate < 10 mEq/L after reposition). After enrollment, patients were randomly assigned to either a higher-intensity (prescribed Kt/V of 0.8 per session) or lower-intensity (prescribed Kt/V of 0.5 per session) PD dose. Randomization was performed using sealed envelopes. The assigned treatment protocol was interrupted when partial recovery of renal function occurred [deined as restoration of diuresis > 1000 mL/24 h, associated with a progressive fall in serum creatinine (<4 mg/dL) and BUN (<50 mg/dL)], when the dialytic method had to be changed because of mechanical complications or infection without treatment success, when renal function was not recovered after 30 days of dialysis, or when the patient died. After the protocol was inter-rupted, living patients were followed for 30 days.
Dialysis dose
The patients in both groups were treated with continu-ous HVPD. A HVPD session was deined as 1 day (24 hours), and sessions were performed 7 days per week. Peritoneal access was established by percutaneous placement of a lexible catheter (Tenckhoff) by a neph-rologists using a trocar introduced paramedian on the left or right periumbilical abdominal wall. Exchanges of 2 L Dianeal (Baxter Healthcare Corporation, Deerield, IL, U.S.A.) PD solution (Na, 135 mEq/L; Ca, 3.5 mEq/L; K, 0 mEq/L; Mg, 1.5 mEq/L; lactate, 40 mEq/L; 1.5% – 4.25% glucose) were performed using an automated HomeChoice cycler (Baxter Healthcare Corporation).
The prescribed HVPD dose was determined us-ing the Kt/V urea formula (12,13), with K beus-ing the volume of dialysis solution prescribed in 24 hours (in milliliters) multiplied by 0.6 (considering that the dialysate/plasma relationship for urea is 0.6 in 1 hour), t is treatment duration (1 day), and V is the urea distribution volume in liters by the Watson formula (14). The prescribed Kt/V value was 0.8 per session for the high-intensity group and 0.5 per session for the low-intensity group. The exchanges used dwell times of 30 – 60 minutes, for a total of 36 – 48 L solution and 18 – 22 exchanges daily. In patients with luid overload, PD solution containing 2.5% or 4.25% glucose was used. Table I shows the characteristics of the PD prescription in each group.
Delivered HVPD dose was determined using the Kt/V urea formula, with K being the mean dialysate/ plasma BUN (in milligrams per 100 mL) before and after dialysis multiplied by the drained volume in
tablei Characteristics of the peritoneal dialysis (PD) sessions
Characteristic Intensity of PD Lower Higher (Kt/V=0.5) (Kt/V=0.8)
Dialysis luid per cycle (L) 2.0 2.0
Inlow time (min) 10 10
Outlow time (min) 20 20
Dwell time (min) 45–60 30–45
Cycle duration (min) 75–90 60–75 Exchanges per session (n) 16–19 20–24
Dialysis luid per session (L) 32–38 40–48
Session duration (h) 24 24
Flow rate (mL/min) 22–26.5 27.8–33.3
milliliters over 24 hours divided by the distribution volume of urea in milliliters (4,13). Blood samples were collected at the beginning and end of each HVPD dialysis session and analyzed for levels of creatinine, potassium, bicarbonate, glucose, and sodium. Three aliquots of spent dialysate (3 mL each) were collected at 8-hour intervals during every session to measure BUN. Dialysate white blood cell count and cultures were also determined every 3 days.
Statistical analysis
The study was designed to demonstrate non-inferiority, and data from at least 60 patients per group were used in calculations, resulting in a statistical power of 80% to detect an absolute difference of 20% in mortality. Mortality was the primary endpoint for calculations; metabolic control was the secondary endpoint used in the design.
The study was prematurely closed because the group assigned to higher-intensity dialysis received a dialysis dose lower than that prescribed. All analyses were performed according to the intention-to-treat principle, with no imputation for missing values. Data from patients lost to follow-up were not analyzed.
Results are presented as medians or means ± stan-dard deviation, according to the normality characteris-tics for each variable, with a 5% (p < 0.05) signiicance level. The independent t-test was used to compare parametric variables between the groups, and for multiple comparisons between the groups, an analysis of variance, followed by a Newman–Keuls test, was used. For nonparametric variables, the Wilcoxon and Kruskal–Wallis tests, followed by the Dunn method, were respectively used to compare two groups and multiple groups. At the end of study, a patient survival curve was calculated for each group.
Statistical analyses were performed, indepen-dently checked, and replicated using the Stata software application (version 8.0, 2004: StataCorp LP, College Station, TX, U.S.A.).
Results
Enrollment
Between January 2005 and January 2007, the study enrolled 79 patients, of whom 39 were assigned to the higher-intensity treatment group, and 40, to the lower-intensity treatment group. Consent was subsequently withheld or withdrawn for 18 patients (23%): 1 patient
required abdominal surgery during the irst week of the study and declined to participate; 9 patients ex-perienced mechanical complications of the peritoneal catheter in the irst 24 hours of dialysis; and 8 patients died during the irst dialysis session. Of the remaining 61 patients, 31 received the higher-intensity PD dose, and 30, the lower-intensity PD dose.
Baseline characteristics
All baseline characteristics were similar between the groups (Table II). Sepsis was the main cause of AKI; heart failure was the second most frequent cause. Sepsis and ischemic acute tubular necrosis were the main forms of AKI; uremic symptoms or azotemia were the main indications for dialysis.
Study and supportive treatments
The groups were similar in metabolic control. Serum levels of BUN, serum creatinine, and bicarbonate stabilized after the same number of dialysis ses-sions. Mean levels after 4 sessions of higher- and lower-intensity therapy were BUN, 112.3 ± 34.4 mg/ dL and 103.8 ± 30.7 mg/dL respectively; creatinine, 3.9 ± 1.1 mg/dL and 3.1 ± 0.9 mg/dL; bicarbonate, 22.7 ± 5.59 mEq/L and 23.9 ± 2.25 mEq/L. Levels of potassium, glucose, and sodium were not signii-cantly different between the groups. Patients from both groups did not present uncontrolled hyperglycemia or hypernatremia during therapy (Figure 1). The deliv-ered dialysis dose was signiicantly higher (38%) in the higher-intensity group than in the lower-intensity group (weekly Kt/V: 4.13 ± 0.6 vs. 3.01±0.5, p = 0.03), but the ultrailtration per session was not sig-niicantly different between the groups (2.41 ± 0.7 L in the higher-intensity group vs. 2.11 ± 0.6 L in the lower-intensity group, p = 0.42).
Ponce et al. 121
The groups showed no signiicant differences in their rates of infectious and mechanical complications. Peritonitis occurred in 12.9% of the patients in the higher-intensity group and in 13.3% of the patients in the lower-intensity group (8 patients in total). With regard to mechanical complications, there was no signiicant difference between the higher- and lower-intensity dose groups (6% vs. 7%, p = 0.41). In both groups, catheter leaks and migration were the main mechanical complications after 24 hours of therapy, but no interruptions in therapy were needed.
Outcome
Overall mortality rates were similar for both groups (55% in the high-intensity group and 53% in the low-intensity group, p = 0.72). Among the 14 high-dose and low-dose patients that survived, 86% in each group recovered renal function after 30 days of therapy
(p = 0.64). Therapy duration was similar in both groups (6.1 ± 2.7 days vs. 5.7 ± 3.1 days, p = 0.42). The analysis of renal function recovery included only survivors, although a few non-survivors recovered par-tial renal function before their deaths. No signiicant difference in actuarial survival was observed between the higher- and lower-intensity PD dose groups (Fig-ure 2). After 30 days of treatment, patient survival was about 50% for both PD doses.
Discussion
The prospective randomized trial reported here was designed to evaluate the intensity of PD therapy in patients with AKI. The main objective was to com-pare mortality rates; the secondary objective was to compare metabolic control.
The groups were similar in their clinical and laboratory characteristics, and although they received tableii Patient characteristics at study enrollment
Characteristic Intensity of PD p
Higher Lower Value
(n=31) (n=30)
Sex (% men) 71 65 0.58
Age (years) 64.2±18.8 62.8±16.2 0.32
Volemiaa (L) 34.8±11.2 34.2±9.6 0.82
Oliguria (%) 56 58 0.78
ATN index-speciic score 0.48
Median 0.67 0.66
Range 0.6–0.76 0.42–0.75
APACHE II score 26.4±6.9 24.8±8.6 0.18
Mechanical ventilation (%) 68 72 0.59
Hemodynamic instability (%) 61 63 0.84
Sessions (n) 0.48
Median 6.1 5.7
Range 4–0.5 5–4
BUN before treatment (mg/100 mL) 118.8±32.6 114.2±34.8 0.78
Cr before treatment (mg/100 mL) 5.61.9 5.8±1.4 0.79
Cause of AKI (%)
Sepsis 48 50 0.77
Heart failure 25 22 0.58
Post surgery 11 14 0.65
Cause of ATN (%)
Ischemic 82 78 0.38
Mixed 16 20 0.61
Indication for dialysis (%)
Uremic symptoms or azotemia 64 66 0.78
Volume overload 20 18 0.84
a Urea distribution in liters by the Watson formula.
Few studies in the literature have compared different dialysis doses in AKI, and the results are mostly conlicting (6–9).
Our indings do not agree with those of three ear-lier randomized controlled studies of continuous RRT in which decreased mortality with increased intensity of treatment was observed. Schifl et al. (18) showed that AKI patients presented fewer fatal complications when receiving daily HD (weekly Kt/V: 5.8) rather than alternate-day HD (weekly Kt/V: 3.8). In a study of 425 patients, Ronco et al. (5) reported a decrease in mortality to 43% from 59% when the prescribed figure 1 Comparison of metabolic control with higher- and
lower-intensity doses of peritoneal dialysis. Median serum levels of (A) blood urea nitrogen (BUN), (B) creatinine, (C) bicarbonate, (D) potassium, (E) glucose, and (F) sodium at the beginning of treatment (“Pre”) and after each session.
statistically different dialysis doses as evaluated by Kt/V, they achieved similar metabolic control after treatment was started. There was no significant difference in mortality rate between the groups, a result that accords with the literature (4,9). In this series, recovery of renal function was similar in both groups, and patient survival after 30 days of treat-ment was not signiicantly different at about 50% in both groups.
The adequacy of the dialysis dose in AKI is a subject of controversy for many reasons (15–18).
tableiii Weekly and per-session urea clearance (Kt/V) and ultrailtration in the study patients
Variable Intensity of PD p Higher Lower Value (n=31) (n=30)
Kt/V Per session
Prescribed 0.8 0.5
Delivered 0.59±0.1a 0.43±0.1 0.03
Weekly
Prescribed 5.6 3.5
Delivered 4.13±0.6a 3.0±0.2 0.03 Ultrailtration (L)
Per session 2.4±0.7 2.1±0.6 0.42
PD = peritoneal dialysis.
figure 2 Comparison of patient survival up to 30 days after
Ponce et al. 123
hourly efluent low was increased from 25 mL/kg to 35 mL/kg or 45 mL/kg. However, the results of our study are similar to those of other randomized controlled studies (9,10,15–18). Tolwani et al. (8) found no difference in outcome among 200 patients randomly assigned to an hourly efluent low of either 20 mL/kg or 35 mL/kg. Our indings are consistent with those of the Acute Renal Failure Trial Network Study (9), which used a combination of continuous and intermittent RRT. In contrast to that study, however, we exclusively used continuous PD as the RRT.
A recent trial by Gabriel et al. (4) showed that PD and daily HD have similar results and suggested that the use of PD was associated with more rapid renal recovery. If the indings of those authors are replicated in large-scale studies, PD may return to the forefront in the management of AKI.
There are several reasons why the use of PD for AKI is declining. The main concern is the possibility of inadequate clearance. A conventional prescription for chronic PD may be inadequate for patients with AKI. Various techniques have been described to achieve higher doses, but no studies had examined the issue of dose in AKI treated with PD. Targets there-fore have to be inferred (10). In the present study, the delivered Kt/V in the higher-intensity group was 38% higher than the delivered Kt/V in the lower-intensity group [similar to results observed in the Ronco et al. study (5)], but no signiicant difference in metabolic control was observed between the higher- and lower-intensity groups. In higher-lower-intensity PD treatment, the weekly delivered Kt/V was lower than the prescribed Kt/V (4.13 vs. 5.6, p = 0.03); in lower-intensity treat-ment, the prescribed and delivered Kt/V were similar (3.01 vs. 3.5, p = 0.59).
In the Acute Renal Failure Trial Network Study (9), the delivered dose for higher-intensity treatment was 89% of the prescribed dose, but Tolwani et al. (16) reported a value of 83%, and the value in the present study was 75%. For lower-intensity treatment, the delivered dose was 95% in the Acute Renal Failure Trial Network Study, 85% in the study by Tolwani et
al., and 86% in the present study.
Phu et al. (19) reported that intermittent PD failed to control acidemia and creatinine levels, which may have led to the higher mortality rate seen in the PD group than in the hemoiltration group. However, those authors used intermittent PD with rigid catheters, a open system, and manual exchanges, and also used
too-short a dwell time, which led to inadequate solute removal. Uchino et al. (2) evaluated two methods of automated PD in AKI patients with mild-to-moderate hypercatabolism. Those authors reported that both methods, continuous PD and tidal PD, were adequate with regard to dialysis dose. Gabriel et al. (3) con-ducted a prospective study with 30 acute renal failure patients who were assigned to high-dose continuous automated PD (Kt/V: 0.65 per session) using a lex-ible PD catheter and a cycler. Patients received 236 continuous PD sessions, for a delivered weekly Kt/V of 3.8 ± 0.6, with metabolic control being achieved after the 4th session of HVPD. Their mortality rate was similar to that observed in the present study.
Based on the characteristics of the peritoneal membrane, PD clearances are known to be limited by dialysate low, membrane permeability, and membrane area (20–22). A 2-L exchange with a dwell time of ap-proximately 30 minutes as proposed by Gabriel et al. (3,4) can achieve a dialysate saturation of about 50%. As a result, over 24 hours, an average urea clearance of approximately 24 L can be expected from a 2 L/h regime. The calculated Kt/V would thus reach 0.6 in a patient with body weight between 65 kg and 70 kg. In the case of a continuous daily regime, the weekly Kt/V would reach 4.2, which is close to the value observed in the present study. If the number of exchanges were to be increased, the dwell time, dialysate saturation, and delivered Kt/V would decrease. It is true that PD is not the most eficient therapy for AKI patients, and clearance per exchange can be decreased if shorter dwell times are applied. Finally, it is not possible to deliver a Kt/V higher than 0.6 with current PD tech-niques. Tidal or continuous lux regimes might achieve a higher Kt/V (21,22).
One of the potential limitations of our study is the small number of patients. The study was performed in a single center and was closed early. We could not deliver the intended dose, because knowledge was lacking about the limitations of PD clearance related to dialysate low, membrane permeability, and membrane area when the study was planned.
Conclusions
practicable and probably will not provide additional clinical beneit. Further studies on the use of PD for AKI, with its associated clinical outcomes, are necessary.
Disclosures
The authors have no inancial conlicts of interest to declare.
References
1 Uchino S, Kellum JA, Bellomo R, et al. Acute renal failure in critically ill patients: a multinational, multi-center study. JAMA 2005;294:813–18.
2 Chitalia VC, Almeida AF, Rai H, et al. Is perito-neal dialysis adequate for hypercatabolic acute renal failure in developing countries? Kidney Int 2002;61:747–57.
3 Gabriel DP, Nascimento GV, Caramori JT, Martim LC, Barretti P, Balbi AL. High volume perito-neal dialysis for acute renal failure. Perit Dial Int 2007;27:277–82.
4 Gabriel DP, Caramori JT, Martim LC, Barretti P, Balbi AL. High volume peritoneal dialysis vs daily hemodialysis: a randomized, controlled trial in pa-tients with acute kidney injury. Kidney Int Suppl 2008;(108):S87–93.
5 Ronco C, Bellomo R, Homel P, et al. Effects of
dif-ferent doses in continuous veno-venous haemoiltra -tion on outcomes of acute renal failure: a prospective randomised trial. Lancet 2000;356:26–30.
6 Saudan P, Niederberger M, De Seigneux S, et al.
Adding a dialysis dose to continuous hemoiltration
increases survival in patients with acute renal failure. Kidney Int 2006;70:1312–17.
7 Bouman CS, Oudemans–Van Straaten HM, Tijssen JG, Zandstra DF, Kesecioglu J. Effects of early
high-volume continuous venovenous hemoiltration on
survival and recovery of renal function in intensive care patients with acute renal failure: a prospective, randomized trial. Crit Care Med 2002;30:2205–11. 8 Tolwani AJ, Campbell RC, Stofan BS, Lai KR, Oster
RA, Wille KM. Standard versus high-dose CVVHDF for ICU-related acute renal failure. J Am Soc Nephrol 2008;19:1233–8.
9 Palevsky PM, Zhang JH, O’Connor TZ, et al. on behalf of the 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.
10 Chionh CY, Soni S, Cruz DN, Ronco C. Peritoneal dialysis for acute kidney injury techniques and dose. Contrib Nephrol 2009;163:278–84.
11 Schrier RW. Nephrology forum: acute renal failure. Kidney Int 1979;15:205–16.
12 Gabriel DP, Nascimento GV, Caramori JT, Martim LC, Barretti P, Balbi AL. Peritoneal dialysis in acute renal failure. Ren Fail 2006;28:451–6.
13 Korbet MS, Kronfo ON. Acute peritoneal dialysis prescription. In: Daugirdas TJ, Blake PG, Ing TS, eds. Handbook of dialysis. 3rd ed. Philadelphia, PA: Lip-pincott Company; 2001: 333–42.
14 Watson PE, Watson ID, Batt RD. Total body water volumes for adult males and females estimated from simple anthropometric measurements. Am J Clin Nutr 1980;33:27–39.
15 Evanson JA, Ikizler TA, Wingard R, et al. Measure-ment of the delivery of dialysis in acute renal failure. Kidney Int 1999;55:1501–8.
16 Paganini EP. Dialysis is not dialysis! Acute dialy-sis is different and needs help! Am J Kidney Dis 1998;32:832–3.
17 Bellomo R, Cass A, Cole L. et al. on behalf of the RENAL Replacement Therapy Study Investiga-tors. Intensity of continuous renal-replacement therapy in critically ill patients. N Engl J Med 2009;361:1627–38.
18 Schifl H, Lang SM, Fischer R. Daily hemodialysis
and the outcome of acute renal failure. N Engl J Med 2002;48:305–10.
19 Phu NH, Hien TT, Mai NT, et al. Hemoiltration and
peritoneal dialysis in infection associated acute renal failure in Vietnam. N Engl J Med 2002;347: 895–902.
20 Ronco C, Amerling R. Continuous low peritoneal di -alysis: current state-of-the-art and obstacles to further development. Contrib Nephrol 2006;150:310–20. 21 Ronco C. Can peritoneal dialysis be considered an
option for the treatment of acute kidney injury? Perit Dial Int 2007;27:251–3.
22 Passadakis PS, Oreopoulos DG. Peritoneal dialysis in patients with acute renal failure. Adv Perit Dial 2007;23:7–16.
Corresponding author:
Daniela Ponce, phd, Distrito de Rubiao Junior,
Botu-catu, Sao Paulo, Brazil. E-mail:
