Published online November 30, 2007
PEDIATRICS
Vol. 120
No. 6
December 2007, pp.
e1380-e1385
(doi:10.1542/peds.2006-3276)
Infant Botulism: A 30-Year Experience Spanning the Introduction of Botulism Immune Globulin Intravenous in the Intensive Care Unit at Childrens Hospital Los Angeles
Karen Underwood, MD,
Sarah Rubin, MD,
Timothy Deakers, MD, PhD and
Christopher Newth, MB
Division of Critical Care, Childrens Hospital Los Angeles, Keck School of Medicine, University of Southern California, Los Angeles, California
 |
ABSTRACT
|
|---|
OBJECTIVE. To report a tertiary care hospital's 30-year experience with the diagnosis, treatment, and outcome of infant botulism in the PICU before and after the availability of Botulism Immune Globulin Intravenous.
METHODS. This was a retrospective medical chart review of the 67 patients who had received a diagnosis of infant botulism and were admitted to the ICU from 1976 to 2005. The ages on presentation, length of hospital stay, length of ICU stay, length of mechanical ventilation, and type of botulism toxin were recorded and compared for patients who had received Botulism Immune Globulin Intravenous and those who had not. On the basis of our results, conclusions were drawn regarding the effect of Botulism Immune Globulin Intravenous on the morbidity of infant botulism.
RESULTS. Sixty-seven patients’ charts were reviewed; 23 male and 29 female patients did not receive Botulism Immune Globulin Intravenous. Of patients who did not receive Botulism Immune Globulin Intravenous, the median age at presentation was 71 days, median length of hospital stay was 35 days, ICU stay was 24 days, and duration of mechanical ventilation was 17 days. A total of 40% had type A toxin, and 60% had type B toxin. There was a significant difference between patients with toxin types A and B in length of hospital stay but not length of ICU stay or mechanical ventilation. Patients with type A toxin were significantly older than patients with type B toxin. Fifteen children received Botulism Immune Globulin Intravenous. There were statistically significant differences in length of hospital stay, length of ICU stay, and length of mechanical ventilation between patients who received Botulism Immune Globulin Intravenous and those who did not.
CONCLUSIONS. The use of Botulism Immune Globulin Intravenous significantly decreased the length of ICU stay, length of mechanical ventilation, and overall hospital stay in children with infant botulism.
Key Words: infant botulism • botulism immune globulin-intravenous • California Department of Health Services
Abbreviations: BIG-IV—Botulism Immune Globulin Intravenous • LOS—length of hospital stay • LOICUS—length of ICU stay • LOMV—length of mechanical ventilation • IQR—interquartile range • CHOP—Children's Hospital of Philadelphia
Approximately 110 cases of botulism in the United States are reported to the Centers for Disease Control and Prevention each year; 72% of reported cases are the infant form of botulism, making this the most prevalent form of botulism in the United States.1 Before 1992, the main treatment for infant botulism was supportive care. A series of trials sponsored by the California Department of Heath Services beginning in 1992 led to the Food and Drug Administration approval of Botulism Immune Globulin Intravenous (BIG-IV) in October 2003. In the past 30 years through the advancement of medical management and the use of BIG-IV, there has been a significant decrease in the morbidity of infant botulism.
Infant botulism is caused by the ingestion of spores that colonize the infant's intestinal tract and produce toxin, whereas foodborne or adult botulism (25% of cases) is caused by the ingestion of preformed botulinum toxin in bacterium-contaminated foods.2 Wound botulism is caused by toxin produced from a wound that is infected with Clostridium botulinum; this is the rarest type of botulism (3%). The highest incidences of infant botulism are in Utah, Pennsylvania, and California.3 It is speculated that the incidence follows these geographic regions because the climate favors their continued existence. These spore-forming bacteria are known to produce 8 different toxin types, and it is known that the distribution of toxin types that cause illness in the United States matches the known distribution of toxin type of botulism-containing spores in the soil (mostly B type toxin east of the Mississippi River and A type toxin west of the Rocky Mountains).4 The general mechanism of action of botulinum toxin is to bind to presynaptic nerve terminals and inhibit the release of acetylcholine, thereby causing muscle paralysis.5 Paralysis is a flaccid, descending phenomenon beginning with the cranial nerves and extending toward the extremities. Symptoms of infant botulism usually begin with constipation and cranial nerve palsies and then progress to generalized weakness. Persistence of these symptoms can result in lengthy hospital stays, admission to the ICU, and need for long-term airway management, which can lead to excessive costs for both families and hospitals.
Approximately 40% of all patients who receive a diagnosis of infant botulism are found in California, with one quarter of these patients residing in Los Angeles County.6 Infant botulism is diagnosed in every month of the year, with peak admissions during the winter months. A sharp increase in admissions (6 cases in 1 month, twice the norm) in the greater Los Angeles area was seen immediately after the Northridge earthquake in January 1994, possibly as a result of spores’ being released into the air. Half of these patients were admitted to our hospital.
The California Department of Health Services created BIG-IV and tested it in a 5-year randomized, double-blind, placebo-controlled statewide trial in California (1992–1997). Subsequently, a 6-year nationwide, open-label study was initiated (1998–2003). Childrens Hospital Los Angeles is a large pediatric tertiary care hospital in Los Angeles County and was a primary participant in both of these studies.7,8 BIG-IV was licensed by the Food and Drug Administration in October 2003 for use for infant botulism,8 and now that BIG-IV is available, Childrens Hospital Los Angeles continues to care for and treat a significant number of patients with infant botulism. This article is a 30-year review of our experience with the diagnosis and treatment of infant botulism in the PICU, with the introduction of BIG-IV occurring close to the midpoint of this period.
|
METHODS
|
|---|
Charts for 75 patients who met the case definition for infant botulism (toxin detected by mouse bioassay in the patient's stool) were found. Of these, 8 had incomplete records. Therefore, 67 cases were included in this retrospective review from 1976 to 2005. Patient demographics, total length of hospital stay (LOS), length of ICU stay (LOICUS), length of mechanical ventilation (LOMV), botulism toxin type, treatment with BIG-IV, disposition at discharge, sequelae, and exposure history were recorded. Only patients who had infant botulism and had been admitted to the ICU were included in our study.
Patients were divided into 4 categories on the basis of their date of admission. Those who were admitted and received a diagnosis before 1990 were designated as Pre-BIG. Those who were admitted and received a diagnosis since 1990 and did not receive BIG-IV were designated as non-BIG. Those who were admitted and received a diagnosis after 1990 and did receive BIG-IV were designated as BIG. The fourth category includes the 52 patients who were admitted and received a diagnosis from 1976 to 2005 and did not receive BIG-IV. Although the BIG-IV study did not start until 1992, we believe that our ICU admission criteria and management changed little during the interval from 1990 to 1992 and roughly equal numbers of patients were placed in the group that received BIG-IV (BIG) and the group that did not (non-BIG).
Student's t test was used for statistical analysis. Logarithmic transformations were used for all variables that were skewed. Statistical analysis was performed by using MedCalc Statistical Software Program (Mariakerke, Belgium).
|
RESULTS
|
|---|
Patients Who Did Not Receive BIG-IV (1976–2005)
Fifty-two patients (23 male, 29 female) were admitted to the ICU for treatment of infant botulism from 1976 to 2005 and did not receive BIG-IV. The median age at presentation for these patients was 71 days (interquartile range [IQR]: 57–102). The median LOS was 35 days (IQR: 30–44), and the median LOICUS was 24 days (IQR: 18–30). A total of 83% of patients with infant botulism required mechanical ventilation. The median LOMV was 17 days (IQR: 9–23). Nine patients did not require mechanical ventilation; 4 of these had type A, and 5 of these had type B. Twenty-one cases were identified as type A (40%), and 31 cases were identified as type B (60%). There was a significant difference between types A and B in LOS (51 days [IQR: 35–71] vs 31 days [IQR: 24–36]; P = .024) but not LOICUS (30 days [IQR: 17–48] vs 20 days [IQR: 12–29]; P = .185) or LOMV (23 days [IQR: 8–36] vs 16 days [IQR: 8–23]; P = .929). Patients with type A were significantly older than patients with type B (121 days [IQR: 69–171] vs 63 days [IQR: 39–80]; P = .0012; Tables 1 and 2).
View this table:
[in this window]
[in a new window]
|
TABLE 1 Comparison of Medians of LOS, LOICUS, LOMV, and Ages of All 67 Patients With Toxin Type A or B Infant Botulism
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 2 Comparison of Medians of LOS, LOICUS, LOMV, and Ages of the 52 Patients Who Did Not Receive BIG-IV With Toxin Type A or B Infant Botulism
|
|
Patients After the Institution of BIG-IV (1990–2005)
Since BIG-IV became available, 15 children received BIG-IV and 12 were ineligible (because of delay in presentation or diagnosis beyond 2 weeks of onset of symptoms) or randomly assigned not to receive it during the same period. There were significant differences in LOS (15 days [IQR: 8–23] vs 27 days [IQR: 13–60]; P = .013), LOICUS (9 days [IQR: 5–13] vs 18 days [IQR: 6–31]; P = .03) and LOMV (6 days [IQR: 2–11] vs 15 days [IQR: 4–35]; P = .03) for patients who received BIG-IV versus those who did not during the same period.
Comparison of Patients Before and After the Introduction of BIG-IV
During the first 15 years of this review, before the use of BIG-IV, a total of 40 patients with infant botulism were admitted to the ICU for treatment. These patients were compared with patients who were admitted to the ICU during the second 15 years of this study (n = 27), after BIG-IV was introduced in California for the treatment of infant botulism. Patients who were admitted before 1990, none of whom received BIG-IV, had a statistically significantly longer total LOS (36 days [IQR: 33–51] vs 15 days [IQR: 8–23]; P = .0001) and LOICUS (27 days [IQR: 19–33] vs 9 days [IQR: 5–13]; P = .0001) than patients who were admitted after 1990 and received BIG-IV. The difference in LOMV between the 2 groups, in contrast, was not statistically significant (18 days [IQR: 9–26] vs 6 days [IQR: 2–11]; P = .69).
We compared the 2 groups of patients who did not receive BIG-IV, including patients before the introduction of BIG-IV (Pre-BIG) and patients who did not receive BIG-IV after it was available (non-BIG). There was a trend toward decreased total LOS (36 vs 27 days), LOICUS (27 vs 18 days), and LOMV (18 vs 15 days) of non-BIG patients compared with Pre-BIG patients, although these results did not reach statistical significance (Table 3). In addition, the proportion of our patients who were intubated did not differ over time or with the advent of BIG-IV. Pre-BIG years included 32 (82%) of 40 intubated and non-BIG years had 11 (95%) of 12. Patients since 1990 who had received BIG-IV (BIG) had 12 (80%) of 15 intubated (Table 3). These differences were not statistically significant.
View this table:
[in this window]
[in a new window]
|
TABLE 3 Comparison of Medians of LOS, LOICUS, LOMV, and Ages of Pre-BIG Patients With Non-BIG or BIG Patients With Toxin Type A or B Infant Botulism
|
|
Two patients received tracheostomies (both before 1985), 1 for subglottic stenosis and 1 for reasons that remain unclear. Other complications in our patients were neither numerous nor severe, with the most frequent being atelectasis and, rarely, pneumonia.
|
DISCUSSION
|
|---|
These results represent the largest number of infants (n = 67) reported with botulism to date and uniquely allow us to observe how the course of this disease was altered with the institution of a new treatment. Childrens Hospital Los Angeles has been following the various toxin types of infant botulism and the outcomes of different treatments since it was first reported in 1976.9 Previously, the most comprehensive study of infant botulism was by Schreiner et al10 at the Children's Hospital of Philadelphia (CHOP) covering the years 1976–1987. This group studied 57 patients with infant botulism, before the availability of BIG-IV, and reported an average age of onset, LOS, LOICUS, and duration of mechanical ventilation slightly different from those in our study for patients who also did not receive BIG-IV during a similar time span. The average ages of our patients and the CHOP patients were similar. Our patients stayed, on average, 10 days longer in the hospital and 11 days longer in the ICU and remained intubated 
7 days longer than the CHOP patients (Table 4); however, only 68% of patients reported by Schreiner et al required mechanical ventilation compared with 82% of our patients. The differences in total LOS, LOICUS, length of intubation, and greater incidence of patients who required mechanical ventilation in our series may reflect the variability in severity of disease, a lower threshold for intubation, or the referral of patients who were more ill at the time of presentation.10 In our institution, we were able to obtain stool samples from every patient and directly identify the organism and/or toxin, making it unnecessary to use other diagnostic criteria such as electromyography as suggested by the CHOP authors. The need for intubation and readiness for extubation of our patients were based on clinical criteria instead of laboratory or pulmonary function tests, the latter having been deemed by Schreiner et al to be difficult and unhelpful. The proportion of our patients who were intubated did not differ over time or with the advent of BIG-IV. During the Pre-BIG years, 82% of patients were intubated; of the non-BIG patients, 95% were intubated; and of patients who received BIG-IV, 80% were intubated.
View this table:
[in this window]
[in a new window]
|
TABLE 4 Comparison of Means of LOS, LOICUS, LOMV, and Age of Patients in the CHOP Study With Patients in the CHLA Study Who Did Not Receive BIG-IV
|
|
Before 1985, 2 patients received tracheotomies. Subsequently, rigorous care was taken with selection of endotracheal tube size, and in no patient to extubation fail for reason of upper airway obstruction from subglottic edema from prolonged intubation. No patients experienced recurrent paralysis after discharge from the ICU or the hospital. All patients survived their hospitalization, and none had any other serious sequelae at the time of discharge. Although all patients received feeds by nasogastric tube, the management of feeding was not specifically addressed in our study.
Significant differences were found between patients with toxin type A and toxin type B infant botulism: patients with type A were significantly older, remained in the hospital longer, and remained intubated longer than patients with type B. These 2 types of botulism have been compared before in other studies and have shown results consistent with ours. Arnon11,12 compared patients with toxin types A and B infant botulism in a study of 50 patients in California and reported an increased LOS of patients with toxin type A versus type B (mean: 39 vs 24 days), although, unlike us, he concluded that there was no significant difference in severity of illness between the 2. Data summarized by Davis13 in 2003 also showed infants with toxin type A botulism to have an increased total LOS over those with toxin type B.
Before 1987, all children with infant botulism were observed in the ICU and treated with supportive care only. Since its first use in clinical trials in 1992 until now, BIG-IV has been used to treat this potentially deadly disease. In the past 13 years, BIG-IV has been shown to be a safe and effective treatment for infant botulism that benefits both patients and parents and by decreasing ICU stay makes valuable resources available for other ICU patients. The main adverse effect of BIG-IV is a rash. Our study showed a 50% decrease in LOS and 50% decrease in LOICUS of patients who received BIG-IV compared with those who did not during the same period. A previous study by Thompson et al3 of 39 patients with infant botulism showed results similar to ours; 13 infants who were treated with BIG-IV had shorter hospital stays than 26 patients who did not receive BIG-IV (23 vs 40 days). In addition, data from the initial BIG-IV study by Arnon4 showed infants who received BIG-IV to have significantly decreased LOS (18.2 vs 39.9 days) and LOMV (4.9 vs 16.8 days) versus those who did not.
When comparing patients who were admitted after BIG-IV began to be widely used with those before BIG-IV, certain trends are evident. Patients who received BIG-IV had shorter total LOS and LOICUS than those before the use of BIG-IV; however, there was also a distinct trend toward shorter hospital stay (Fig 1), ICU stay (Fig 2), and LOMV (Fig 3) between the historical control subjects and the patients who were randomly assigned not to receive BIG-IV after its introduction. As is often the case with historical control subjects, the 2 populations of patients being compared may be quite different. Before BIG-IV, all patients who received a diagnosis of infant botulism were taken care of in the ICU, so the LOICUS actually reflected the total LOS. This may have caused the LOICUS to be more prolonged than they might have been if patients could have been admitted outside the ICU. Perhaps patients were more ill on presentation before BIG-IV, because less was known about infant botulism and patients received the diagnosis later in the disease process. On the basis of these trends, it is clear that after 1990, changes occurred, either in the way patients with botulism were cared for or in the disease itself. BIG-IV has been shown to reduce mean hospital costs by approximately $100000 per case by decreasing LOS and need for ICU management.8 After subtracting the cost of BIG-IV (approximately $45000) from the total savings, $55000 still remains as savings to the hospital.
View larger version (23K):
[in this window]
[in a new window]
|
FIGURE 1 Comparison of mean LOS for patients with infant botulism from 1976 to 1989 before the use of BIG-IV (Pre-BIG) to patients from 1990 to 2005 who did not receive BIG-IV (Non-BIG) or did receive BIG-IV (BIG).
|
|
View larger version (9K):
[in this window]
[in a new window]
|
FIGURE 2 Comparison of the logarithmic transformations of the LOICUS for patients with infant botulism from 1976 to 1989 before the use of BIG-IV (Pre-BIG) to patients with infant botulism from 1990 to 2005 who did not receive BIG-IV (Non-BIG) or did receive BIG-IV (BIG). a P = not significant when non-BIG compared with pre-BIG; b P = .03 BIG compared with non-BIG; c P = <.001 BIG compared with pre-BIG.
|
|
View larger version (8K):
[in this window]
[in a new window]
|
FIGURE 3 Comparison of LOMV of patients after 1990 who did not receive BIG-IV (Non-BIG) with those who did receive BIG-IV (BIG). a P = .03 BIG compared with non-BIG.
|
|
|
CONCLUSIONS
|
|---|
Giving BIG-IV as soon as the diagnosis of infant botulism is suspected can decrease the morbidity and expense of this life-threatening disease. Treatment with BIG-IV should be initiated promptly and not delayed for laboratory confirmation of diagnosis.14 Current recommendations are that it not be administered if the onset of symptoms of infant botulism is > 10 days before diagnosis.
The clinical focus in infant botulism should be on maintaining a high index of suspicion for this disease and on recognizing symptoms early to institute appropriate therapy, including BIG-IV, at the earliest time; however, our experience over 30 years shows that with or without BIG-IV therapy, the ICU complications of patients with infant botulism, even for those who require prolonged intubation, are modest, and there have been no mortalities.
|
ACKNOWLEDGMENTS
|
|---|
We thank Drs Wendy Mitchell and Greg Pacentine for greatly appreciated contributions to this project. Thanks also go to Prof Frank Shann for review and statistical advice and to Dr Stephen Arnon for many contributions to our care of these patients and guidance through the BIG-IV project.
|
FOOTNOTES
|
|---|
Accepted May 16, 2007.
Address correspondence to Sarah Rubin, MD, Childrens Hospital Los Angeles, Anesthesiology and Critical Care, 4650 Sunset Blvd, Los Angeles, CA 90027. E-mail sarrubin{at}chla.usc.edu
The authors have indicated they have no financial relationships relevant to this article to disclose.
|
REFERENCES
|
|---|
- Centers for Disease Control and Prevention.
Botulism. Atlanta, GA: Centers for Disease Control and Prevention; 2001. Available at: www.cdc.gov/ncidod/dbmd/diseaseinfo/botulism-g.htm. Accessed October 1, 2006
- American Academy of Pediatrics. Clostridial infections. In: Pickering LK, Baker CJ, Long SS, McMillan JA, eds.
Red Book: 2006 Report of the Committee on Infectious Diseases. 27th ed. Elk Grove Village, IL: American Academy of Pediatrics; 2006:257
–260
- Thompson JA, Filloux FM, Van Orman CB, et al. Infant botulism in the age of botulism immune globulin.
Neurology. 2005;64
:2029
–2032[Abstract/Free Full Text]
- Arnon SS. Infant botulism. In: Feigin RD, Cherry JD, Demmler GJ, Kaplan SL, eds.
Textbook of Pediatric Infectious Diseases. 5th ed. Vol 2. Philadelphia, PA: WB Saunders; 2003;1758
–1766
- Samuelson J. Infectious diseases. In: Cotran R, Kumar V, Collins T, eds.
Robbins Pathologic Basis of Disease. 6th ed. Philadelphia, PA: WB Saunders; 1999:368
–369
- California Department of Health Services. Infant botulism-cases and costs-1985–1994. Biweekly Report from the Division of Communicable Disease Control; 1995: June 30. Available at: www.dhs.ca.gov/ps/dcdc/cm/950630cm.htm. Accessed October 1, 2006
- Arnon SS.
Clinical Trial of Human Botulism Immune Globulin: Botulism and Tetanus Neurotoxins. New York, NY: Plenum; 1993:477
–482
- Arnon SA, Schechter R, Maslanka SE, Jewell NP, Hatheway CL. Human botulism immune globulin for the treatment of infant botulism.
N Engl J Med. 2006;354
:462
–471[Abstract/Free Full Text]
- Midura TF, Arnon SS. Infant botulism: identification of Clostridium botulinum and its toxins in faeces.
Lancet. 1976;2(7992)
:934
–935
- Schreiner MS, Field E, Ruddy R. Infant botulism: a review of 12 years’ experience at the Children's Hospital of Philadelphia.
Pediatrics. 1991;87
:159
–165[Abstract/Free Full Text]
- Arnon SS. Infant botulism treatment and prevention program: open-label administration of Human Botulism Immune Globulin. Available at: www.dhs.ca.gov/dcdc/infantbot/openlbl.htm. Accessed October 1, 2006
- Arnon SS. Infant botulism.
Annu Rev Med. 1980;31
:541
–560[CrossRef][ISI][Medline]
- Davis L. Botulism.
Curr Treat Options Neurol.5
:23
–31, 2003[Medline]
- Francisco AM, Arnon SS. Clinical mimics of infant botulism.
Pediatrics. 2007;119
:826
–828[Abstract/Free Full Text]
PEDIATRICS (ISSN 1098-4275). ©2007 by the American Academy of Pediatrics
This article has been cited by other articles:
|
|
|
|
 
W. G. Mitchell and L. Tseng-Ong
Reviews of Infant Botulism at Childrens Hospital Los Angeles
J Child Neurol,
August 1, 2008;
23(8):
968 - 968.
[PDF]
|
|
|
|
|
|
|
M. H. Rathore
BIG Babies Do Better with Infant Botulism
AAP Grand Rounds,
March 1, 2008;
19(3):
30 - 31.
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
METHODS
