Rapid Responses to:

PAPERS: Urs Eriksson, Burkhard Seifert, and Andreas Schaffner Comparison of effects of amphotericin B deoxycholate infused over 4 or 24 hours: randomised controlled trial BMJ 2001; 322: 579 [Abstract] [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Complications of amphotericin therapy

Mike Dennis   (23 March 2001)

In response to Dr. Dennis

Eriksson Schaffner   (9 April 2001)

Adverse effects of rapid versus slow infusion of amphotericin B

Julio Collazos   (28 June 2001)

Complications of amphotericin therapy 23 March 2001
Mike Dennis, specialist regisrtar Dept. Haematology, University of Liverpool. Send response to journal: Re: Complications of amphotericin therapy	EDITOR- Eriksson et al report a comparative assessment of infusion rates for amphotericin B 1, concluding that continuous infusions lead to reduced nephrotoxicity and other adverse side effects with at least comparable efficacy. Unusually, only approximately half the patients included in the study had a fever in the twenty-four hours prior to commencing therapy. Again only a similar number appeared to be receiving concurrent antibiotic therapy. Both these features are at odds with the experience within our own unit or that of previously published trials 2. Whilst amphotericin may be started for pulmonary infiltrates suggestive of aspergillus, fungal colonisation or positive microbiological cultures the vast majority are for fever unresponsive to multiple broad spectrum antibiotics. The conclusive findings are welcomed though they make little reference to the impracticality of such a protocol. The use of intravenous amphotericin B is largely confined to neutropenic patients undergoing chemotherapy for haematological malignancies. While these patients routinely have dual lumen central venous catheters, such as hickman lines. After allowance for regular blood sampling, fluid/electrolyte replacement, parenteral nutrition, antibiotics and chemotherapy regimes there is frequently inadequate access for the conventional rapid infusion of amphotericin B. Positioning of further access for continuous infusion would unfortunately be unacceptable to most patients (and some staff) with the risk of an additional source for sepsis. Mike Dennis, Specialist Registrar in Haematology Department of Haematology, Royal Liverpool University Hospital, Liverpool, L69 3GA m.dennis@liverpool,ac.uk 1 Eriksson U, Seifert B, Schaffner A. Comparison of effects of amphotericin B deoxycholate infused over 4 or 24 hours: randomised controlled trial. BMJ 2001;322:579-582. 2 Prentice HG, Hann IM, Herbrecht R, Gibson BES et al. A randomised comparison of liposomal versus conventional amphotericin B for the treatment of pyrexial of unknown origin in neutropenic patients. Brit. J. Haem 1997; 98: 711-718
In response to Dr. Dennis 9 April 2001
Eriksson Schaffner Zurich University Hospital Send response to journal: Re: In response to Dr. Dennis	Dr Dennis makes welcome comments in regard to our randomized study comparing the effect of the infusion rate on the toxicity of amphotericin B deoxycholate. The fraction of neutropenic patients receiving antibiotics when starting amphotericin B was 100%, table 2 gives only the numbers for the treatment with nephrotoxic agents (aminogycosides and vancomycin). As our centre is plagued with pulmonary aspergillosis in high-risk patients we have adopted an aggressive strategy towards pulmonary mold infection. This strategy is reflected in the definitions of refractory fever and possible, probable and proved fungal infections given in the paper. Patients with any chest symptoms (cough, chest pain) undergo a chest CT scan or are started on amphotericin B until such a radiological evaluation can be performed. New infiltrates typical for a mold infection are a clear-cut indication for antifungal therapy in our epidemiological setting. The indications to start amphotericin B in the studied population are given in our table 2. While microbiological examinations of respiratory secretions are not performed routinely at our centre visualization of molds in bronchial secretions in a high risk patient with granulocyte counts < 100/mm3 would prompt us to start antifungal therapy. We agree with Dr. Dennis that the practical aspects of continuous infusions of amphotericin B are very important. We routinely use antimicrobially coated triple lumen catheters at our centre and have not changed this policy for the purpose of our study. Patients see no difference between the insertion of a double and triple lumen catheter and we never encountered problems in giving in parallel to a continuous infusion of amphotericin B, antibiotics, chemotherapy, blood products and parenteral nutrition. The insertion of an additional i.v- line was never required. The nursing staff is pleased to have no longer to deal with shivering and nauseated patients requiring additional medications and was glad to adopt continuous infusion of amphotericin B throughout our hospital. By using this approach we can reassure Dr. Dennis that patient and staff satisfaction will be high and the pharmacy bill will stay low. Andreas Schaffner, MD Professor of Medicine, University of Zurich Urs Eriksson, MD Medical ICU, University of Basel
Adverse effects of rapid versus slow infusion of amphotericin B 28 June 2001
Julio Collazos, Section of Infectious Diseases Hospital de Galdakao, 48960 Vizcaya, Spain Send response to journal: Re: Adverse effects of rapid versus slow infusion of amphotericin B	In their recent article Eriksson et al [1] analysed the side effects related to the infusion of amphotericin B deoxycholate over 4 or 24 hours. They concluded that the slower infusion rate was associated with lower incidence of side effects. Although the only infusion-related reactions specified in their report were fever, chills or rigors, vomiting and headache, more severe reactions may also occur associated with the infusion of this drug. Recently, we cared for a patient who developed a fatal pulmonary oedema during the infusion of amphotericin B lipid complex. To elucidate this poorly understood condition we reviewed the published cases of pulmonary reactions to amphotericin B [2]. Acute pulmonary events have been described during the infusion of all formulations of amphotericin B. Many of these events were classified as anaphylactic, although we believe that there are many arguments that suggest that the mechanisms of these reactions are not allergic but chemically mediated. Characteristically these reactions occur during the first days of treatment and, in most cases, shortly after the first dose of the drug. Besides chest symptoms, other infusion-related complaints such as fever and chills were commonly observed. Pulmonary reactions seemed to be closely related to the infusion rate. Several reactions developed upon the switch from amphotericin B deoxycholate to the lipid formulations of amphotericin B, which represented considerably higher doses administered in substantially shorter infusion periods. Interestingly, pulmonary events recurred when the offending drug was readministered at the same infusion rates, but they did not recur in any patient in whom amphotericin infusion was slowed. Although rapid infusion rates have been reported to cause similar infusion-related events than slower ones in some studies [3-5], they were characterised by important limitations that could have had a major influence in their results [2]. Of note, our patient tolerated the infusion of amphotericin lipid complex over 6 hours for two days, but developed the severe pulmonary reaction on the third day when the infusion period was reduced to two hours. This observation, as well as the other reported cases of pulmonary reactions to different formulations of amphotericin B, support the findings of Eriksson et al [1] that slower infusion rates of amphotericin are safer than rapid ones. We believe that amphotericin B should be infused more slowly than usually recommended during the first few days of treatment, when the incidence and intensity of infusion-related events are higher, and that shorter infusion periods may be safely administered later if desired. Julio Collazos, Eduardo Martínez, Jose Mayo, Sofía Ibarra Section of Infectious Diseases, Hospital de Galdakao, Vizcaya, Spain Correspondence: Dr. J Collazos, Section of Infectious Diseases, Hospital de Galdakao, 48960 Vizcaya, Spain. E- mail: jcollazos@hgda.osakidetza.net References 1. Eriksson U, Seifert B, Schaffner A. Comparison of effects of amphotericin B deoxycholate infused over 4 or 24 hours: randomised controlled trial. BMJ 2001;322:579-582. 2. Collazos J, Martínez E, Mayo J, Ibarra S. Pulmonary reactions during treatment with amphotericin B. Review of the published cases and guidelines for management. Clin Infect Dis (in press). 3. Cruz JM, Peacock JEJ, Loomer L, Holder LW, Evans GW, Powell BL, et al. Rapid intravenous infusion of amphotericin B: a pilot study. Am J Med 1992;93:123-130 4. Nicholl TA, Nimmo CR, Shepherd JD, Phillips P, Jewesson PJ. Amphotericin B infusion-related toxicity: comparison of two- and four-hour infusions. Ann Pharmacother 1995;29:1081-1087. 5. Cleary JD, Weisdorf D, Fletcher CV. Effect of infusion rate on amphotericin B-associated febrile reactions. Drug Intell Clin Pharm 1988;22:769-772.