|
|
| |||||||||||||||||||||||||||||||||||||||||
Background Atrial natriuretic peptide, a hormone synthesized by the cardiac atria, increases the glomerular filtration rate by dilating afferent arterioles while constricting efferent arterioles. It has been shown to improve glomerular filtration, urinary output, and renal histopathology in laboratory animals with acute renal dysfunction. Anaritide is a 25-amino-acid synthetic form of atrial natriuretic peptide.
Methods We conducted a multicenter, randomized, double-blind, placebo-controlled clinical trial of anaritide in 504 critically ill patients with acute tubular necrosis. The patients received a 24-hour intravenous infusion of either anaritide (0.2 µg per kilogram of body weight per minute) or placebo. The primary end point was dialysis-free survival for 21 days after treatment. Other end points included the need for dialysis, changes in the serum creatinine concentration, and mortality.
Results The rate of dialysis-free survival was 47 percent in the placebo group and 43 percent in the anaritide group (P = 0.35). In the prospectively defined subgroup of 120 patients with oliguria (urinary output, <400 ml per day), dialysis-free survival was 8 percent in the placebo group (5 of 60 patients) and 27 percent in the anaritide group (16 of 60 patients, P = 0.008). Anaritide-treated patients with oliguria who no longer had oliguria after treatment benefited the most. Conversely, among the 378 patients without oliguria, dialysis-free survival was 59 percent in the placebo group (116 of 195 patients) and 48 percent in the anaritide group (88 of 183 patients, P = 0.03).
Conclusions The administration of anaritide did not improve the overall rate of dialysis-free survival in critically ill patients with acute tubular necrosis. However, anaritide may improve dialysis-free survival in patients with oliguria and may worsen it in patients without oliguria who have acute tubular necrosis.
Methods
Patients
Patients at least 18 years of age with a clinical diagnosis of acute tubular necrosis due to recent ischemic or nephrotoxic insults were eligible for enrollment. The diagnosis of acute tubular necrosis was based on the patient's clinical history, physical examination, and laboratory values, including an analysis of urinary electrolytes and examination of urine sediment. Patients were also required to have a continuing increase in serum creatinine of at least 1 mg per deciliter (88 µmol per liter) in a period of less than 48 hours despite the optimization of fluid status. Patients were excluded from the study if they had acute renal dysfunction due to prerenal azotemia, vascular obstruction, postrenal obstruction, or systemic or intrinsic renal diseases other than acute tubular necrosis (such as vasculitis and glomerulonephritis); if they had already undergone dialysis for the current episode of acute tubular necrosis; if they were expected to require dialysis within the subsequent 24 hours or not to be candidates for dialysis; or if their underlying nonrenal medical condition was so severe that an improvement in renal function would not be expected to improve the clinical outcome. We also excluded patients who had a history of marked chronic renal insufficiency (usual serum creatinine concentration [before acute tubular necrosis], >3.0 mg per deciliter [265 µmol per liter]), previous renal transplantation, or a systolic blood pressure of less than 90 mm Hg despite the use of vasopressor therapy. The study protocol was approved by the institutional review board at each center, and all the patients gave informed consent.
Assignment and Administration of the Study Drug
To enroll a patient, a member of the staff at the study site telephoned an independent Study Randomization and Interim Analysis Center that randomly assigned eligible patients in a double-blind manner in a 1:1 ratio, stratified according to center, to receive an infusion of either anaritide (Auriculin, or human atrial natriuretic peptide, amino acid residues 102 to 126; Scios, Mountain View, Calif.) or an identical-appearing placebo (stored in coded vials). The infusion of anaritide (or an equal volume of placebo) was initially given at a dose of 0.05 µg per kilogram of body weight per minute intravenously. The dose was escalated to 0.20 µg per kilogram per minute over a 90-minute period and was continued at that level (or at the highest dose the patient tolerated) for the remainder of the 24-hour treatment period.
Study Protocol
Each patient's blood pressure and heart rate were monitored, and single-lead electrocardiography was performed, during the infusion of anaritide or placebo. A complete blood count and tests of serum chemistry were done before and after the administration of the drug. Urinary output and creatinine clearance were measured for at least four hours before the administration of anaritide or placebo; urinary output was measured for five days after treatment. Serum creatinine concentrations, requirements for dialysis, and mortality were followed for 21 days, and dialysis status and mortality were assessed again at day 60.
All the patients received full supportive care for their acute renal failure, including the optimization of fluid and nutritional status and necessary treatment for other medical problems, including the adjustment of doses of medication as appropriate for patients with renal dysfunction. Low-dose dopamine (<3 µg per kilogram per minute) or diuretic therapy was instituted or continued at the discretion of the investigator. The need for dialysis was determined by each patient's attending nephrologist on a case-by-case basis, with the need judged on the basis of the presence of volume overload, electrolyte imbalance, uremia, or acid–base disturbances not responsive to medical management.
Statistical Analysis
The primary end points were dialysis-free survival (the percentage of patients who survived through day 21 without requiring dialysis) and dialysis (the percentage of patients who underwent dialysis by day 14). Dialysis-free survival was analyzed by the Pearson chi-square test. Dialysis was studied by first estimating the curves for the time to the first dialysis with Kaplan–Meier methods, with death used as a censoring mechanism. We then used the point estimates at day 14 to test the null hypothesis that the incidence of dialysis was the same in both groups; Greenwood's formula was used to estimate the variances. All statistical tests were two-sided. Plus–minus values are means ±SD.
Secondary end points included mortality from any cause and changes in the serum creatinine concentration at day 21. Analyses of multiple subgroups were prespecified in the protocol and were defined by the absence or presence of multiorgan failure, the cause of acute tubular necrosis (nephrotoxic or ischemic insult), the serum creatinine concentration at enrollment (<4.0 mg per deciliter [354 µmol per liter] vs. >4.0 mg per deciliter), the presence or absence of a history of chronic renal insufficiency (usual serum creatinine concentration, <1.8 mg per deciliter [159 µmol per liter] vs. 1.8 to 3.0 mg per deciliter), and urinary output at enrollment (<100 ml per day vs. 100 to 399 ml per day [both considered oliguric] vs. >400 ml per day [nonoliguric]).
An independent Data Monitoring Committee reviewed issues of study execution and safety on an ongoing basis and conducted three interim analyses, in which data on 151, 294, and 366 patients were summarized. The objectives of the interim analyses included evaluating safety and determining whether there was sufficient evidence of efficacy or lack thereof to warrant an early termination of the study; any such decision was to be based on a prespecified guideline for stopping. At no time during the study did either the investigators or the sponsor have access to the randomization code or the unblinded results of the interim analyses. The final analysis was performed on an intention-to-treat basis, with SAS version 6.08, run on a VAX 4000-105A; it included all patients enrolled in the study.
Results
Between January 1993 and February 1995, 504 patients with acute tubular necrosis were enrolled at 59 clinical centers in the United States and Canada. One patient in the placebo group was lost to follow-up on day 11 without having undergone dialysis. All the other patients were followed through day 21 or death.
Characteristics of the Patients
At enrollment, 425 patients (84 percent) were in the intensive care unit, and 253 patients (50 percent) were intubated for respiratory support. Acute tubular necrosis was attributed primarily to a nephrotoxic insult in 112 patients (22 percent), to an ischemic insult in 132 patients (26 percent), and to multiple causes in 255 patients (51 percent). The treatment groups were well matched with regard to base-line demographic variables, medical history, and the severity of renal dysfunction at enrollment (Table 1).
|
During the infusion of the study drug, 167 patients (33 percent) received no dopamine, 170 patients (34 percent) received low-dose dopamine, and 167 patients (33 percent) received higher-dose dopamine for hemodynamic support; 303 patients (60 percent) received diuretics. The proportions treated with diuretics and dopamine were similar in the anaritide and placebo groups (data not shown). Concomitant diuretic therapy did not alter the responses to anaritide. There was a trend toward better outcomes in the patients with oliguria who received anaritide but not dopamine than in those who received anaritide plus dopamine. Because the patients were not randomly assigned to receive dopamine therapy, however, these findings could reflect the differential administration of dopamine to patients with a worse prognosis.
Among patients who underwent dialysis, there was no significant difference in the type of dialysis membrane used (that is, a cellulose membrane vs. a noncellulose membrane), either between the oliguric and nonoliguric groups or between the anaritide and placebo groups. Serum creatinine and urea nitrogen values before the start of dialysis were also similar in the anaritide and placebo groups.
Measurements of Outcome
The rates of dialysis-free survival for 21 days in the anaritide and placebo groups were 43 and 47 percent, respectively (P = 0.35). By day 14, 42 percent of the patients in the placebo group and 44 percent of those in the anaritide group had undergone dialysis (P = 0.75). The rates of death from any cause by day 21 were 26 percent in the placebo group (67 of 256 patients) and 29 percent in the anaritide group (73 of 248 patients, P = 0.41). The mean serum creatinine concentration at day 21 in the placebo group was 3.0±2.2 mg per deciliter (265±194 µmol per liter); in the anaritide group, it was 2.8±2.0 mg per deciliter (248±177 µmol per liter, P = 0.98).
Analyses of Subgroups
Table 2 shows rates of dialysis-free survival according to treatment group in the overall study population and all prospectively defined subgroups. Among the patients with oliguria, dialysis-free survival was 8 percent in the placebo group (5 of 60 patients) and 27 percent in the anaritide group (16 of 60 patients, P = 0.008). Conversely, among the patients without oliguria, dialysis-free survival was 59 percent in the placebo group (116 of 195 patients) and 48 percent in the anaritide group (88 of 183 patients, P = 0.03). When the study population was stratified further according to base-line urinary output, there was a trend toward improved dialysis-free survival after the administration of anaritide in the subgroups with base-line urinary output of less than 400 ml per day, whereas in the subgroups with base-line urinary output of at least 400 ml per day, rates of dialysis-free survival tended to decrease (Figure 1).
|
|
Among the 120 patients with oliguria, 87 percent of those in the placebo group but only 64 percent of those in the anaritide group underwent dialysis by day 14 (P = 0.005). Among the patients without oliguria, 30 percent of those in the placebo group and 38 percent of those in the anaritide group underwent dialysis (P = 0.12). The mean serum creatinine concentration at day 21 was lower in the patients with oliguria who received anaritide than in those who received placebo (2.8±1.9 mg per milliliter [248±168 µmol per liter] vs. 4.1±2.6 mg per milliliter [362±230 µmol per liter], P = 0.06).
There were no significant treatment-related differences in survival to day 60 in either the oliguric group or the nonoliguric group (Figure 2). Among the patients with oliguria, the rates of death from any cause by day 21 were 45 percent in the placebo group (27 of 60 patients) and 40 percent in the anaritide group (24 of 60 patients, P = 0.58). Among the patients without oliguria, the corresponding rates were 20 percent in the placebo group (39 of 195) and 26 percent in the anaritide group (48 of 183, P = 0.15).
|
The administration of anaritide was generally tolerated well. Hypotension was reported in 18 percent of the placebo group and 46 percent of the anaritide group (P<0.001). The absolute and proportional declines in blood pressure during the infusion of anaritide were greater in patients who did not have oliguria than in those who did. Premature ventricular contractions during the infusion were noted in 6 percent of patients in the anaritide group and 2 percent of those in the placebo group (P = 0.009); serious arrhythmias (such as heart block, ventricular tachycardia, and cardiac arrest) were infrequent (in 3 percent of patients or less) and were evenly distributed between the treatment groups. No other major adverse events and no adverse changes in laboratory values were found in the anaritide group.
Discussion
We found that a 24-hour intravenous infusion of anaritide did not improve overall survival without dialysis in critically ill patients with acute tubular necrosis. The results of the subgroup analysis, however, suggest that anaritide infusion may have different effects in such patients according to their base-line urinary output, improving dialysis-free survival in those with oliguria but perhaps worsening it in those without oliguria.
The findings of subgroup analyses must always be interpreted with caution,15 particularly when multiple subgroups are tested, because the likelihood of false positive results is increased. To assess the possibility that the results in patients with oliguria were an artifact caused by the testing of multiple subgroups, we ran a computer simulation (data not shown). In each iteration, subgroup membership was randomly reassigned and the outcomes in the subgroups were reanalyzed. Only 15 of 5000 iterations (0.3 percent) yielded a result for the subgroup that was similar to the actual findings for the patients with oliguria in this trial, suggesting that these results would only rarely be obtained by chance.
The differing effects of anaritide treatment on dialysis-free survival according to base-line urinary output in patients with acute tubular necrosis were accompanied by differential trends in subsequent renal function (for example, in the serum creatinine concentration and urinary output), the need for dialysis, and mortality. Parallel positive trends in outcome were also found in prospectively defined subgroups of patients with oliguria (that is, patients with a urinary output below 100 ml per day and those with an output of 100 to 399 ml per day). Among patients with oliguria, conversion to nonoliguric status after treatment with anaritide — but not placebo — correlated with improvement in dialysis-free survival.
Blood pressure during the infusion of anaritide decreased more in the patients who did not have oliguria than in those who did. This may have resulted in decreased renal blood flow and further ischemic injury in the patients without oliguria, thus worsening renal function.16,17 This finding is consistent with the observation that during the anaritide infusion the mean urinary output of the patients with oliguria increased, whereas in those without oliguria it decreased.
The differential responses to anaritide according to the level of urinary output may also reflect intrinsic differences between patients with oliguria and those without it with regard to the intrarenal mechanisms that regulate renal dysfunction. In a recent study of the effects of the biocompatibility of the dialysis membrane on the clinical outcome in patients with acute renal failure, the use of polymethyl methacrylate membranes as compared with cuprophane membranes resulted in improved recovery of renal function and a trend toward decreased mortality in patients who did not have oliguria, but not in those who did.18 Thus, among patients with acute tubular necrosis, the extent of intrarenal vasoconstriction, immune-mediated disease, and glomerular and tubular dysfunction and the effectiveness of the kidney's regenerative capabilities may differ between patients with oliguria and those without it. The intrarenal pharmacologic actions of anaritide may improve renal function in patients with oliguria but may have no benefit — or may even have deleterious effects — in patients without oliguria.
Our findings are consistent with those of earlier studies indicating that patients with acute tubular necrosis who have oliguria have worse clinical outcomes than those who do not,1,2 and they suggest that the two groups of patients may respond differently to treatment with drugs such as anaritide. In future clinical studies of patients with acute tubular necrosis, consideration should be given to stratifying patients prospectively on the basis of their urinary output.
Supported by Scios, Inc. Drs. Allgren and Genter are employees of Scios, Inc.
We are indebted to the study coordinators and to the other study personnel at all the participating sites for their assistance in recruiting patients and conducting this study.
Source Information
From the Clinical Research Division, Scios, Inc., Mountain View, Calif. (R.L.A., F.C.G.); Orlando Clinical Research Center, Orlando, Fla. (T.C.M.); University of Texas Health Science Center, Houston (S.N.R.); Cooper Hospital, Camden, N.J. (L.S.W., B.R.C.K.); Baylor University Medical Center, Dallas (A.Z.F.); New England Medical Center, Boston (R.A.L.); Kidney Disease and Critical Care Associates, Minneapolis (R.M.S.); Denver Veterans Affairs Medical Center, Denver (J.D.C.); and Brigham and Women's Hospital, Boston (M.H.S.). Presented as an abstract at the 28th Annual Meeting of the American Society of Nephrology, San Diego, Calif., Nov. 5–8, 1995, and the 13th International Congress of Nephrology, Madrid, Spain, July 2–6, 1995.
Address reprint requests to Dr. Sayegh at the Renal Division, Brigham and Women's Hospital, 75 Francis St., Boston, MA 02115.
References
Study Randomization and Interim Analysis Center — Maryland Medical Research Institute: G. Knatterud, K. Ra, S. Forman, M. Terrin. Data Monitoring Committee: S. Klahr (chair), T. Fleming, L. Hunsicker, H. Smith.
The following additional institutions and investigators participated in this study. Brigham and Women's Hospital, Boston: G.M. Chertow, J.M. Lazarus; Oregon Health Sciences University, Portland: W.M. Bennett, M. Meyer; University of Texas Health Science Center, Houston: A.R. Butt, T.D. DuBose; South Florida Nephrology Associates, Lauderdale Lakes: E.R. Martin; Wake Forest University, Winston-Salem, N.C.: P. Adams, A. Tuttle; Medical Center of Delaware, Wilmington: W. Miller, D. Maichle; Nephrology Medical Associates, Tarzana, Calif.: K. Kleinman, G.E. Fischmann, S. Schweitzer; St. Joseph's Hospital, Tampa, Fla.: R.J. Goldstein, J.O. Navarro; Cleveland Clinic Foundation, Cleveland: R.J. Heyka; Duke University Medical Center, Durham, N.C.: S.J. Schwab, P. Conlon; Toledo Hospital, Toledo, Ohio: K. Lempert, G. Haig; Lenox Hill Hospital, New York: M.F. Michelis, R. Chan, M.V. DeVita; Vanderbilt University Medical Center, Nashville: J. Breyer, G. Schulman; University of California San Diego Medical Center, San Diego: R. Mehta; University of Arizona Health Sciences Center, Tucson: Y.H. Lien; Roudebush Veterans Affairs Medical Center, Indianapolis: J.A. Hasbargen, A. O'Shaughnessy; Montefiore Medical Center, Bronx, N.Y.: N. Bank, M. Coco; UCLA Medical Center, Los Angeles: A. Wilkinson; Veterans Affairs Medical Center, Dayton, Ohio: M. Saklayen; Wilford Hall Medical Center, Lackland Air Force Base, Tex.: D. Burgess; Medical College of Virginia, Richmond: T.W.B. Gehr; Austin Diagnostics Clinic, Austin, Tex.: B. Welch; Henry Ford Hospital, Detroit: M. Faber, P. Abrego; University Hospital, Denver: A. Merouani; Temple University Hospital, Philadelphia: M. Goldberg; University of Ottawa, Ottawa, Ont., Canada: P. Magner; University of Utah, Salt Lake City: F. Shihab; Deaconess Hospital, Boston: R. Solomon; University of Alabama–Birmingham, Birmingham: D. Warnock; University of Alberta, Edmonton, Alta., Canada: C. Kjellstrand, W. Chin, R. Huizinga; University of Texas Health Science Center, San Antonio: C. Nolan; St. John's Mercy Medical Center, St. Louis: M. Ravenscraft; Wayne State University, Detroit: J. Sondheimer, H. Dries, J. Bandes; University of Texas Southwestern Medical Center, Dallas: R. Toto, J. Smart; Clinical Research Associates of Tidewater, Norfolk, Va.: D. Wombolt; Rhode Island Hospital, Providence: R. Endreny, M. Maher; University of Michigan, Ann Arbor: E. Young; St. Boniface General Hospital, Winnipeg, Man., Canada: A. Fine; San Francisco General Hospital, San Francisco: M. Humphreys; Tulane Medical Center, New Orleans: J. Puschett, S. DiLeo; West Roxbury Veterans Affairs Medical Center, West Roxbury, Mass.: G. Curhan; Brookdale Hospital Medical Center, Brooklyn, N.Y.: P. Faubert; State University of New York Health Sciences Center, Brooklyn, N.Y.: E. Friedman; Emory University School of Medicine, Atlanta: J.M. Sands; St. Louis University Health Sciences Center, St. Louis: K. Martin; University of Chicago Medical Center, Chicago: J. Umans; University of California at Irvine Medical Center, Orange: N.D. Vaziri, C. Kaupke; and Mt. Sinai Medical Center, Cleveland: T. Zipp.
| |||||||||||||||||||||||||||||||||||||||||
Related Letters:
Acute Oliguria
Goodkin D. A., Narins R. G., Merin R. G., Dalton R.G., Pope J., Klahr S., Miller S. B.
Extract |
Full Text
N Engl J Med 1998;
339:201-202, Jul 16, 1998.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
