Current issues of ACP Journal Club are published in Annals of Internal Medicine


HA-1A reduced in-hospital mortality in gram-negative bacteremia with septic shock, but not in all patients with sepsis

ACP J Club. 1991 May-June;114:72. doi:10.7326/ACPJC-1991-114-3-072

Source Citation

Ziegler EJ, Fisher CJ, Sprung CL, et al. Treatment of gram-negative bacteremia and septic shock with HA-1A human monoclonal antibody against endotoxin. A randomized, double-blind, placebo-controlled trial. N Engl J Med. 1991;324:429-36.



To evaluate the effect of treatment with HA-1A (Centoxin, a human monoclonal IgM antibody) on gram-negative bacteremia and the sepsis syndrome.


Randomized, double-blind, placebo-controlled trial of 28 days duration.


24 academic medical centers in the United States, Canada, and Europe.


543 patients ≥ 18 years of age who had a body temperature of > 38.3 °C or < 35.6 °C; tachycardia and tachypnea; and either hypotension or signs of systemic toxicity or peripheral hypoperfusion were included. Patients were ineligible if the condition was clearly irreversible or was caused by uncontrolled hemorrhage, cardiogenic shock, or infection associated with burn injury; if pregnancy was suspected; or if the patient had received monoclonal antibodies within 21 days. 99% were followed for 28 days.


HA-1A, 100 mg, diluted with 3.5 g of human serum albumin (n = 262) (or serum albumin alone as placebo, n = 281) was given intravenously for 15 to 20 minutes. Other treatment was determined on an individual basis.

Main outcome measure

In-hospital mortality among patients who proved to have gram-negative bacteremia.

Main results

At randomization, 200 patients (37%) had gram-negative bacteremia isolated from blood cultures (105 HA-1A recipients and 95 placebo recipients). 32 (30%) of the 105 HA-1A recipients died compared with 45 (49%) of the 92 placebo recipients still being followed at 28 days (P = 0.014). {This absolute risk reduction (ARR) of 19%, means that 5 patients with gram-negative bacteremia would need to be teated (NNT) with HA-1A (compared with placebo) to prevent 1 additional death within 28 days of treatment, 95% CI 3 to 21; the relative risk reduction (RRR) was 38%; CI 11% to 57%.}* Among patients who were in shock before the infusion, 18 of 54 HA-1A recipients (33%) died compared with 27 of 47 placebo recipients (57%) (P = 0.017). {ARR 24%; NNT 4, CI 2 to 21; RRR 42%, CI 10% to 63%.}* HA-1A was effective in reducing mortality independently of severity of illness, other clinical variables, and adequacy of other therapy for septicemia. There was no difference in mortality between treatment groups among patients without gram-negative bacteremia (P = 0.68) or when results for all patients were analyzed (P > 0.24).


HA-1A human monoclonal antibody was effective in reducing in-hospital mortality among septic patients with gram-negative bacteremia, but had no effect on patients with septicemia from other causes.

Source of funding: Centocor, Inc.

Address for article reprint: Dr. C.J. Fisher, Center for Critical Care Research, University Hospitals of Cleveland, 2074 Abington Road, Cleveland, OH 44106, USA.

*Numbers calculated from data in article.


Despite aggressive supportive therapy and appropriate use of antibiotics, gram-negative sepsis, with or without shock, continues to cause substantial mortality in hospitalized patients. Better treatments are certainly needed. Encouraging results were reported several years ago using immunotherapy with human antibody directed against a determinant expressed by the J5 mutant of Escherichia coli (1). This new study by Ziegler and colleagues further supports the use of immunotherapy, showing a reduction in mortality among patients with gram-negative sepsis.

Because therapy was initiated before the diagnosis of gram-negative bacteremia was confirmed by blood culture, an overall benefit for all cases of sepsis was not observed. However, the reduction in mortality was clinically and statistically significant in those who were later confirmed to have gram-negative bacteremia. Thus, the authors feel that the reduction in mortality and the low risk for side effects warrant the use of HA-1A therapy before culture results are available in patients for whom there is high suspicion of gram-negative infection and who meet well-defined criteria for sepsis. Using these criteria, fewer than 40% of patients proved to have gram-negative sepsis and thus a chance for benefit. Research is underway to develop tests for earlier detection of endotoxemia and gram-negative bacteremia, but no tests are yet suitable for bedside diagnosis.

At present, HA-1A is under active review by the U.S. Federal Drug Administration, and the manufacturer hopes for approval for prescription use this year.

Sharon Peters, MD
The General HospitalSt. John's, Newfoundland, Canada


1. Ziegler EJ, McCutchan JA, Fierer J, et al. Treatment of gram-negative bacteremia and shock with human antiserum to a mutant Escherichia coli. N Engl J Med. 1982;307:1225-30.