Correspondence
Hemofiltration and Peritoneal Dialysis in Infection-Associated Acute Renal Failure
N Engl J Med 2003; 348:858-860February 27, 2003
- Article
To the Editor:
The study by Phu et al. (Sept. 19 issue)1 demonstrated that peritoneal dialysis is associated with a poorer outcome than hemofiltration in the treatment of infection-associated acute renal failure. We believe that certain factors in addition to the one discussed by Daugirdas in the accompanying editorial2 may account for this poorer outcome.
The very short dwell time — a half-hour — may account for the poor clearance, leading to higher creatinine levels and slower correction of acidosis, and may explain the considerably poorer outcome in the patients who underwent peritoneal dialysis. The volume of dialysate used may have been high for the generally small patients in the study, possibly leading to abdominal distention and causing problems with ventilation and gas exchange. The use of stiff catheters and a large volume of dialysate may also have allowed leakage, which, in turn, could have increased the risk of peritonitis. The rate of culture-negative peritonitis was very high (whereas usually it is less than 20 percent), a situation that could have led to undertreatment. It is noted that the dialysate was cloudy in only a small proportion of the cases. This may be because the dwell time was too short for cloudiness to develop in the fluid. The current guidelines of the International Society for Peritoneal Dialysis recommend that, in equivocal cases, a second exchange should be performed with a dwell time of two hours.3
All these factors must be considered before peritoneal dialysis is deemed inappropriate for infection-associated renal failure. Successful use of this technique in a similar setting has recently been reported.4
4 ReferencesPanduranga S. Rao, M.D., D.N.B.
Kulwant Singh Modi, M.D.
Medical College of Ohio, Toledo, OH 43614
prao@mco.edu 1
Phu NH ,Hien TT ,Mai NTH , et al. Hemofiltration and peritoneal dialysis in infection-associated acute renal failure in Vietnam. N Engl J Med 2002;347:895-902
Full Text | Web of Science | Medline2
Daugirdas JT . Peritoneal dialysis in acute renal failure -- why the bad outcome? N Engl J Med 2002;347:933-935
Full Text | Web of Science | Medline3
Keane WF ,Bailie GR ,Boeschoten E , et al. Adult peritoneal dialysis-related peritonitis treatment recommendations: 2000 update. Perit Dial Int 2000;20:396-411[Erratum, Perit Dial Int 2000;20:828-9.]
Web of Science | Medline4
Chitalia VC ,Almeida AF ,Rai H , et al. Is peritoneal dialysis adequate for hypercatabolic acute renal failure in developing countries? Kidney Int 2002;61:747-757
CrossRef | Web of Science | Medline
To the Editor:
Phu et al. show that continuous hemofiltration with 25 liters of lactate-based replacement fluid per 24-hour period is superior to peritoneal dialysis with 70 liters of acetate-based fluid per 24-hour period (with a 30-minute dwell time) with respect to the clearance of solute and the correction of acidemia. Even given the difference between the randomized groups in solute clearance, the difference between the groups in survival cannot be confidently attributed to the assigned method of dialysis.
The challenge in studies of acute renal failure has been the recruitment of sufficient numbers of patients in the randomized groups to prevent imbalance in important clinical predictors of survival. It is clear from previous, larger studies that adequate randomization protocols have failed to produce balanced groups with respect to major variables such as the Acute Physiology and Chronic Health Evaluation score,1-3 sex,2 age,1 the presence or absence of sepsis,3,4 and the degree of organ failure.2 Despite multivariate analysis in these trials, failure of randomization with respect to important variables raises the question of residual confounding by unmeasured or unknown but clinically important variables. Adjustment for underlying disease and jaundice status may not adequately control for important clinical variables; details of coexisting conditions, the severity of illness, and other active interventions were not captured by Phu et al.
To date, randomized trials involving patients with acute renal failure have been underpowered to assess survival outcomes. This study, along with previous studies, can provide insight into the assessment of this outcome for future trials.
4 ReferencesLiam F. Casserly, M.B.
Boston University Medical Center, Boston, MA 02118
lfc@bu.edu 1
Jorres A ,Gahl GM ,Dobis C , et al. Haemodialysis-membrane biocompatibility and mortality of patients with dialysis-dependent acute renal failure: a prospective randomised multicentre trial. Lancet 1999;354:1337-1341
CrossRef | Web of Science | Medline2
Mehta RL ,McDonald B ,Gabbai FB , et al. A randomized clinical trial of continuous versus intermittent dialysis for acute renal failure. Kidney Int 2001;60:1154-1163
CrossRef | Web of Science | Medline3
Ronco C ,Bellomo R ,Homel P , et al. Effects of different doses in continuous veno-venous haemofiltration on outcomes of acute renal failure: a prospective randomised trial. Lancet 2000;356:26-30
CrossRef | Web of Science | Medline4
Schiffl H ,Lang SM ,Fischer R . Daily hemodialysis and the outcome of acute renal failure. N Engl J Med 2002;346:305-310
Full Text | Web of Science | Medline
To the Editor:
The conclusion reached by Phu et al. — that the method of renal-replacement therapy accounts for the observed difference in mortality among patients with infection-related renal failure — is misleading. The direct cause of death in 41 percent of the patients was hemodynamic shock, a condition that cannot be ascribed directly to renal failure. Rather, renal dysfunction in these patients was probably initiated by decreased renal perfusion (prerenal azotemia) secondary to overwhelming infection. Clinical data pertinent to the magnitude of prerenal causes such as the hemodynamic status at base line or laboratory values such as the urea nitrogen–creatinine ratio in plasma or the renal-failure index1,2 are, unfortunately, not provided. The rate of survival would improve with better control of the primary illness through timely antibiotic therapy, combined with proper hemodynamic support, rather than as a result of the method of renal-replacement therapy chosen.
2 ReferencesOren Fruchter, M.D.
National Institutes of Health, Bethesda, MD 20892-1583
fruchteo@mail.nih. gov 1
Brady HR, Brenner BM, Clarkson MR, Lieberthal W. Acute renal failure. In: Brenner BM, ed. Brenner & Rector's the kidney. 6th ed. Vol. 2. Philadelphia: W.B. Saunders, 2000:1201-62.
2
Brady HR, Brenner BM. Acute renal failure. In: Braunwald E, Fauci AS, Kasper DL, Hauser SL, Longo DL, Jameson JL, eds. Harrison's principles of internal medicine. 15th ed. Vol. 2. New York: McGraw-Hill, 2001:1541-51.
To the Editor:
What is impressive in the report by Phu et al. is the difference in mortality associated with the use of peritoneal dialysis in patients with acute renal failure due to malaria and sepsis. In the accompanying editorial, Daugirdas speculates on the factors that may have contributed to the difference in mortality between the two groups, including the difference in rates of clearance. Other factors that may have contributed include peritonitis, the use of acetate, hyperglycemia, and high splanchnic glucose levels, promoting the growth of organisms in malaria.
The median time to death in the peritoneal-dialysis group was 29.8 hours, as compared with 13.8 hours in the hemofiltration group. The majority of the deaths were related to severe acidosis and prolonged hemodynamic shock, leading to cardiorespiratory arrest. It is hard to imagine that any of the above factors could lead to a difference in mortality of this magnitude. However, the difference between the two techniques is that peritoneal dialysis involves the instillation of 2 liters of fluid into the peritoneal cavity. This technique can lead to increased intraabdominal pressure, decreased venous return, decreased ventilation, and decreased organ perfusion. This combination of events can lead to hemodynamic compromise. These factors may have caused prolonged hypotension and severe acidosis, leading to death.
Hasan Bazari, M.D.
Massachusetts General Hospital, Boston, MA 02114
hbazari@partners.org Author/Editor Response
We disagree with Drs. Rao and Modi, who suggest that the outcome in the peritoneal-dialysis group could have been improved substantially by modification of our peritoneal-dialysis technique. Rapid exchanges with a dwell time of 30 minutes optimize ultrafiltration and are more efficient at clearing urea and other solutes than more protracted exchanges.1,2 The widespread use of modern machines for acute peritoneal dialysis in pediatric nephrology is a testament to the efficacy of this approach. We agree that the presence of large volumes of dialysis fluid in the abdomen may have adverse consequences, particularly in terms of ventilation. This is an inherent problem with all peritoneal-dialysis techniques. We do not believe that the volumes used in our study were excessive. Although 2 liters was the standard volume of dialysate instilled per cycle, the volume varied according to the size of the patient.3
The large difference in mortality between the two treatment groups probably did not result from unequal randomization, as suggested by Dr. Casserly. We compared 23 base-line demographic, clinical, and laboratory features and indications for dialysis, of which 10 have been shown in previous studies to be statistically associated with the risk of death in patients with malaria, bacterial sepsis, or both; none of these variables differed significantly between the two groups. All the patients had established acute renal failure with low blood urea nitrogen–creatinine ratios. Most of the patients were previously fit young adults, so coexisting illness was not a major confounder, as distinct from the patient populations in intensive care units in the developed world. In addition, the two groups were not imbalanced with respect to antibiotic or antimalarial treatment or with respect to the availability and appropriate use of inotropic support.
We stated clearly that the study was not designed to assess mortality as the primary end point, and that the observed difference in survival may therefore overestimate the true difference, since estimates obtained when clinical trials are stopped early tend to be biased from zero. However, the difference in mortality is supported by very large differences in the rates of the primary end point: the correction of biochemical and acid–base abnormalities. Acidosis was an important cause of death (in 73 percent of those who died); it was corrected within 24 hours in the majority of patients in the hemofiltration group (but not in the peritoneal-dialysis group). This probably influenced the outcome. It is this combined evidence that leads us to conclude that hemofiltration is preferable to peritoneal dialysis for the treatment of severe infection-associated acute renal failure.
3 ReferencesNguyen Hoan Phu, M.D.
Tran Tinh Hien, M.D.
Hospital for Tropical Diseases, Ho Chi Minh City, VietnamNicholas Day, B.M., B.Ch.
John Radcliffe Hospital, Oxford OX3 9DU, United Kingdom
nick.day@ndm. ox. ac. uk 1
Boen ST . Kinetics of peritoneal dialysis: a comparison with the artificial kidney. Medicine (Baltimore) 1961;40:243-287
CrossRef | Web of Science2
Sorkin MI, Diaz-Buxo JA. Physiology of peritoneal dialysis. In: Daugirdas JT, Ing TS, eds. Handbook of dialysis. 2nd ed. Boston: Little, Brown, 1994:245-61.
3
Trang TT ,Phu NH ,Vinh H , et al. Acute renal failure in patients with severe falciparum malaria. Clin Infect Dis 1992;15:874-880
CrossRef | Web of Science | Medline
- Citing Articles (1)
Citing Articles
1
Yoh Hirayama, Hiroyuki Hirasawa, Shigeto Oda, Hidetoshi Shiga, Kazuya Nakanishi, Kenichi Matsuda, Masataka Nakamura, Takeshi Hirano, Takeshi Moriguchi, Eizo Watanabe, Masakazu Nitta, Ryuzo Abe, Takaaki Nakada. (2003) The Change in Renal Replacement Therapy on Acute Renal Failure in a General Intensive Care Unit in a University Hospital and its Clinical Efficacy: A Japanese Experience. Therapeutic Apheresis and Dialysis 7:5, 475-482
CrossRef
