6 February 2002

Dear Editors

May et al. have provided another instructive case with further useful comments on the management of infant botulism. In reply to their request for details of the neurophysiology findings, we would happily provide anyone interested with a copy of the complete report, but will add the following summary of relevant details for comparison with other possible future cases.

The nerve conduction velocities were borderline low/normal and latencies normal. The amplitudes were all reduced. Repetitive stimulation resulted in no increment at 2, 20 or 50 Hz, nor was any decrement seen at 2 Hz. The electromyogram showed insertional activity and, at rest, profuse fibrillation-like potentials. On effort, the compound motor action potentials were of short duration. The picture was consistent with denervation, as in infant botulism, but equally with other anterior horn cell pathologies and many other diseases.

Typical electromyography (EMG) findings in infant botulism are not specific to the illness; therefore, EMG cannot serve as a definitive diagnostic study. The brief, small, abundant motor-unit potentials often observed in infant botulism, which may also be found in other conditions, were explained and given the acronym BSAP by W King Engle at the National Institutes of Health in the USA many years ago.1

Definitive diagnosis in our patient was made by the mouse neutralization assay, which identified botulinum toxin type A in the faecal extract; Clostridium botulinum was also cultured from the faeces. None of the honey samples yielded organisms resembling C. botulinum. We agree with May et al. regarding the need for early stool samples in order to establish the diagnosis. The inability to identify toxin in faeces in their case was almost certainly the consequence of the 1-month interval between hospital admission and diagnostic testing of faeces, as excretion of toxin and organisms falls off in the weeks and months after onset, until no longer detectable (which is further evidence that C. botulinum is not part of the normal resident flora of humans).2 Not very much stool is needed to demonstrate the presence of toxin; additional quantity (mass) does not improve the yield. If a first stool specimen is inadequate for whatever reason, a second one promptly obtained usually resolves the situation. Testing of stool for C. botulinum toxin and organisms remains the best diagnostic investigation, while other laboratory tests that May et al. mention, such as enzyme-linked immunosorbent assay and polymerase chain reaction are not available or clinically validated.

May et al. describe the clinical variation in the duration of illness. Their patient was hospitalized for 17 weeks. This length of stay is rare, but not unknown. In the pre-Botulism Immune Globulin (BIG) era in California (1976−1991), approximately 1.5% of more than 500 patients had stays of 17 weeks or more (SS Arnon pers. comm.). While it may be correct that ‘most published cases have admissions between 26 and 37 days in length . . .’, this simply reflects that fact that fewer individual cases are published. The mean hospital stay in the placebo group of the pivotal clinical trial of human BIG was 5.5 weeks (i.e. a substantial number of patients had to have had hospital stays exceeding 37 days).3 The effect of aminoglycosides in prolonging convalescence has been previously described.4

The particular value in publishing the clinical details of these cases is also to allow earlier diagnosis if clinicians are aware of how the condition presents. We agree with the advice against equine-derived botulinum immune globulin (i.e. botulinum antitoxin), and clarification of guidelines from the Technical Advisory Group on Immunisation would be appropriate. Information on the management of infant botulism is available in the chapter in the Textbook of Pediatric Infectious Diseases,3 the Red Book5 and on the Internet at the Infant Botulism Treatment and Prevention Program site (California Department of Health Services; Outside the USA, BIG is unavailable in accordance with USA federal law. The Orphan Drug Act 1983 requires that an orphan drug sponsor make available at all times a supply of the product sufficient to meet the needs of the USA; however, there is not a sufficient supply of product that can meet these needs as well as the global demand. Although any patient hospitalized in the USA is eligible to receive BIG, most physicians and parents would agree that a trans-oceanic transfer of a critically ill infant is not in the best interests of the patient.


  1. Top of page
  2. References
  • 1
    Engel WK. Brief, small, abundant motor-unit action potentials: a further critique of electromyographic interpretation. Neurology 1975; 25: 1736.
  • 2
    Arnon SS, Midura TF, Damus K, Snowden S, Thompson B, Chin J. Persistent fecal excretion of Clostridium Botulinum toxin and organisms in patients recovering from infant botulism. In: Proceedings of the Annual Meeting of the American Society for Microbiology[Abstract C(H)67]. American Society for Microbiology, Los Angeles, CA, 1979; 357.
  • 3
    Arnon SS. Infant botulism. In: Feigin RD, Cherry JD, eds. Textbook of Pediatric Infectious Diseases, 4th edn. WB Saunders, Philadelphia, 1998; 157077.
  • 4
    L’Hommedieu C, Stough R, Brown L, Kettrick R, Polin R. ­Potentiation of neuromuscular weakness in infant botulism by aminoglycosides. J. Pediatr. 1979; 95: 106570.
  • 5
    American Academy of Pediatrics. Botulism in Red Book, Report of the Committee on Infectious Diseases, American Academy of Pediatrics, 25th edn. American Academy of Pediatrics, Elk Grove Village, IL, 2000.