 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
© 1996 The Journal of Bone and Joint Surgery, Inc. Musculoskeletal Complications of Varicella*   , SAN DIEGO, CALIFORNIAInvestigation performed at the Childrens' Hospital, San Diego
Varicella, commonly known as chickenpox, is a common viral infection in children. An estimated 3.5 million cases occur annually in the United States. Serious musculoskeletal complications such as osteomyelitis and necrotizing fasciitis, although uncommon, can be life and limb-threatening. The purpose of the current study was to determine the association between varicella and serious musculoskeletal complications necessitating operative treatment and to characterize these infections in otherwise healthy children. We retrospectively reviewed the records of all patients who had been admitted to the Children's Hospital in San Diego because of varicella and its complications during the eleven-year period from 1984 through 1994. All records with an inpatient discharge diagnosis code for varicella were identified. Twenty-seven (6 per cent) of the 417 admissions for varicella were for musculoskeletal complications of the disease necessitating operative treatment. There were seven admissions for osteomyelitis, four for septic arthritis, five for necrotizing fasciitis, ten for a deep-tissue abscess, and one for toxic shock syndrome leading to multiple limb amputations. Seventy-nine (19 per cent) of the 417 admissions occurred in 1994. Eleven (41 per cent) of the twenty-seven musculoskeletal complications that led to operative treatment occurred in 1994, representing a significant increase in the number of such complications compared with the numbers in previous years of the study (p < 0.01). Bacterial pathogens were identified as the cause of twenty-five of the twenty-seven complications that led to operative treatment. Of these twenty-five, twenty-one (84 per cent) were found, on culture, to be caused by group-A ß-hemolytic streptococcus. This pathogen was the cause of the infection in five of the seven patients who had osteomyelitis while Staphylococcus aureus was the cause in only one. Group-A ßhemolytic streptococcus was also the causative organism in two of the four patients who had septic arthritis, three of the five who had necrotizing fasciitis, and all ten who had a deep-tissue abscess. Nine of the eleven musculoskeletal complications leading to operative treatment in 1994 had group-A ß-hemolytic streptococcus as the causative organism. An understanding of the trends of and a high level of suspicion for potentially serious secondary infections in children who have varicella is necessary for prompt recognition and appropriate treatment.
Varicella is a common viral infection in children, with an estimated 3.5 million cases occurring annually in the United States22. Serious complications, although uncommon, can be life and limb-threatening. Osteomyelitis2,5,9,11-13,17,19,20,31, necrotizing fasciitis2,3,7,24,30, septic arthritis1,4,8-10,26,27,31, and abscesses14 have been reported in children who have varicella. However, we know of no large studies in which the frequency of musculoskeletal infection necessitating operative intervention has been characterized or documented. The purpose of the current study was to determine the association between varicella and serious secondary musculoskeletal infections necessitating operative treatment in a large cohort of children. We also sought to characterize the etiology of such infections and to determine the contribution of group-A ß-hemolytic streptococcus to these trends in disease.
We performed a retrospective review of the records of all patients who had been admitted to Children's Hospital in San Diego because of a diagnosis of varicella between January 1, 1984, and December 31, 1994. The records of these patients were identified through the use of discharge diagnosis coding. Of the 472 admissions that were identified, fifty-five were excluded because the primary reason for admission was unrelated to complications of varicella. Thus, 412 patients (417 admissions) formed the basis of this study. Each record was reviewed with regard to the patient's age, the medical history including immunocompetency, the onset of the varicella rash, and the diagnosis at admission and discharge. Physical and laboratory findings, the results of cultures of blood and specimens from the site of infection, imaging studies, details of treatment, operative findings, results of pathological analysis of operative specimens, and outcome were recorded for all patients who had musculoskeletal complications necessitating operative intervention. To determine if there was a significant trend or a change in the hospital population, census data were obtained for the years of the study. A chi-square test was performed to analyze the number of musculoskeletal complications relative to the total number of hospital admissions and to the number of admissions for all complications related to varicella in 1994 as compared with previous years of the study.
The 417 admissions were grouped according to the discharge diagnosis (Fig. 1). Twenty-seven admissions (twenty-six patients) were for musculoskeletal complications of varicella that led to an operation. All twenty-six patients were immunocompetent. The average age of the patients was 4.1 years (range, four months to eleven years). Fourteen patients were girls and twelve were boys.
There were ten admissions because of a deep-tissue abscess, seven because of osteomyelitis (which was associated with septic arthritis in three patients), five because of necrotizing fasciitis, four because of isolated septic arthritis, and one because of toxic shock leading to multiple limb amputations (Table I). There were no deaths in this series. The average time from the onset of the rash to admission (excluding the second admission for one patient, Case 23) was six days (range, two to thirteen days). No apparent relationship was found between the reported subjective severity of the varicella rash (the number of lesions) and the occurrence of musculoskeletal complications.
Fourteen (54 per cent) of the twenty-six patients had musculoskeletal complications involving the lower extremities, nine (35 per cent) had involvement of the back or the trunk, and six (23 per cent) had involvement of the upper extremities. Three complications involved both the upper and lower extremities. All five patients who had necrotizing fasciitis had involvement of the back or the trunk. Four of the ten deep-tissue abscesses were in the lower extremities and one was in an upper extremity; the remaining five were in the trunk or the neck. Osteomyelitis most commonly involved the distal part of the femur (three patients), followed by the humerus and the tibia (two patients each). Septic arthritis most commonly involved the knee (five patients), followed by the elbow (two patients) (Table I). Two patients (Cases 5 and 16) had multiple foci of osteomyelitis and simultaneous involvement of at least one adjacent joint. One patient (Case 9) had an infected ankle joint adjacent to an isolated focus of osteomyelitis. An additional patient (Case 23) was initially diagnosed with a deep soft-tissue abscess in the thigh as well as cellulitis of the cephalad part of the chest and the neck. Initially, there was no evidence of osteomyelitis; however, the patient was seen again six weeks later with clinical and radiographic evidence of osteomyelitis in the distal part of the contralateral femur, and all cultures were positive for group-A ß-hemolytic streptococcus. Another patient (Case 10) had streptococcal toxic-shock syndrome with gangrene secondary to vascular compromise and infection in all four extremities, and amputations involving all four limbs had to be performed. Of the twenty-seven admissions for musculoskeletal complications, twenty-five necessitated at least one operative procedure in the operating room. Two patients had a procedure performed at the bedside for culture and drainage. The complications led to an average of 1.7 operative procedures (range, one to seven procedures) per patient. The average time from admission to the first operative procedure was two days (range, zero to nine days). The initial débridement for the five patients who had necrotizing fasciitis was performed within twenty-four hours (three patients) or on the sixth or seventh day of hospitalization (two patients). For the latter two patients, a delay in recognition of the extent of the infection led to a delay in operative treatment. All patients were managed with antibiotics intravenously; four patients also received acyclovir intravenously. Of the 417 admissions, seventy-nine (19 per cent) were in 1994. Despite a steady yearly increase in the hospital population due to a growing patient base, the ratio of admissions for varicella-related complications to total hospital admissions increased significantly in 1994 compared with the ratios in previous years of the study (p < 0.01) (Table II). Furthermore, eleven (41 per cent) of the twenty-seven admissions for musculoskeletal complications that led to operative treatment were in 1994 (Fig. 2). The ratio of admissions for musculoskeletal complications to admissions for all complications related to varicella also increased significantly in 1994 (p < 0.01).
The causative bacterial pathogen was identified for twenty-five of the twenty-seven infections. Group-A ß-hemolytic streptococcus grew on culture of specimens from the sites of twenty-one infections (78 per cent). Notably, in five of the seven patients who had osteomyelitis the causative organism was group-A ß-hemolytic streptococcus, while Staphylococcus aureus was the causative organism in only one patient who had osteomyelitis. Group-A ß-hemolytic streptococcus was the causative organism in three of the five patients who had necrotizing fasciitis, two of the four who had septic arthritis, all ten who had a deep-tissue abscess, and the one patient who had toxic shock syndrome leading to multiple limb amputations (Fig. 3). Nine of the musculoskeletal complications leading to operative treatment in 1994 had group-A ß-hemolytic streptococcus as the causative organism.
Four patients had bacteremia; in three, this was caused by group-A ß-hemolytic streptococcus. Four patients had clinical shock; in only one (Case 10) was bacteremia with group-A ß-hemolytic streptococcus documented. As for the three remaining patients who had shock, group-A ß-hemolytic streptococcus grew on cultures of specimens from the site of the infection only (Table I).
Musculoskeletal complications of varicella have been described in the literature5 since 1935 however, there have been few case reports or large studies of these complications. Osteomyelitis is an uncommon complication of varicella. The cases of thirteen patients who had osteomyelitis associated with varicella were found in the literature2,5,9,11-13,17,19,20,31. In nine of these cases, a bacterial etiology was reported, with group-A ß-hemolytic streptococcus cited as the causative organism in seven and Staphylococcus, in two. In the current series, five of the seven patients who had osteomyelitis had group-A ß-hemolytic streptococcus as the causative organism. This is notable, given that the most common etiology of osteomyelitis in children who do not have varicella is Staphylococcus aureus25. Necrotizing fasciitis as a complication of varicella has also been described by several authors. The largest such series (fourteen children) of which we are aware was reported on recently by Brogan et al.3. Although the duration of follow-up was only two years, the number of patients is notable given that only eight other previous cases were found in the literature2,7,24,30. All of their patients had group-A ß-hemolytic streptococcus as the etiological organism, a finding that was similar to our observations. Furthermore, as noted earlier, a delay in the diagnosis led to a delay in treatment for many of their patients. Septic arthritis as a complication of varicella has been reported in nine patients1,4,8-10,26,27,31. The causative organism was group-A ß-hemolytic streptococcus in six patients and Staphylococcus aureus in three. In the current series, four patients had isolated septic arthritis (three in the knee and one in the elbow). Additionally, three of our patients with osteomyelitis had septic arthritis in at least one adjacent joint, and two of the three had polyfocal osteomyelitis. To our knowledge, septic arthritis in association with osteomyelitis after varicella has not been reported previously in the literature. Five of the seven patients in our series who had septic arthritis had cultures that were positive for group-A ß-hemolytic streptococcus; one, cultures that were positive for Staphylococcus aureus; and one, cultures that remained negative. The latter patient was thought to have an infectious process despite the negative cultures because of the clinical characteristics, which included purulent-appearing synovial fluid with a white-blood-cell count of 142,000 per cubic millimeter (14.2 x 109 per liter) and 88 per cent polymorphonuclear leukocytes. As in the current series, a deep soft-tissue abscess was the most common musculoskeletal complication in the large study of 2534 children (both outpatients and inpatients) who had varicella reported by Bullowa and Wishik5 in 1935. Those authors noted twenty-four abscesses; six drained spontaneously, seventeen needed operative drainage, and the treatment of one was not reported. The predominant organism was hemolytic streptococcus5. All ten patients in our series who had a deep soft-tissue abscess had cultures that were positive for group-A ß-hemolytic streptococcus. The literature reflects little variation in the etiology of secondary infectious complications of varicella over time. As mentioned, Bullowa and Wishik5 cited hemolytic streptococcus as the primary bacterial cause of secondary infection. More recently, Jackson et al.16 reported an increase, after 1985, in the number of varicella-related soft-tissue infections in extremities that were caused by group-A ß-hemolytic streptococcus. The notable exception appears to be the series of Fleisher et al.10, in which Staphylococcus aureus was the cause of secondary soft-tissue bacterial infection in most patients. Our study documents a strong association between group-A ß-hemolytic streptococcus and musculoskeletal complications of varicella during an eleven-year period. Group-A ß-hemolytic streptococcus may have a selective advantage in patients who have varicella because of its ability to gain access to deeper tissue through the varicella vesicle itself. The vesicle is created by a split in the epidermis that fills with a proteinaceous fluid15. This may provide a route from the surface of the skin to the subcutaneous tissues, with subsequent bacteremia, or local spread, and musculoskeletal invasion. Trauma to the skin, such as that from scratching, could contribute to bacterial contamination of the varicella vesicle. Group-A ß-hemolytic streptococcus possesses tissue-dissolving enzymes such as hyaluronidase and streptolysin, which facilitate penetration of the tissue18. Although particular strains of group-A ß-hemolytic streptococcus have been reported to express virulence factors (M proteins 1 and 3) and exotoxins (streptococcal pyogenic exotoxins A, B, and C) associated with invasive disease28, we could not address the prevalence of such strains or virulence factors because of the retrospective nature of our study. In the current study, twenty-seven (6 per cent) varicella-related complications that necessitated hospitalization were musculoskeletal infections that led to operative treatment. The Centers for Disease Control23 has estimated that there are 177 hospital admissions for varicella and its complications for every 100,000 children who have varicella before the age of fifteen years. On the basis of these data, we calculate an approximate rate of one musculoskeletal complication for every 10,000 children who have varicella. In the final year of our study, the number of patients who had musculoskeletal complications secondary to group-A ß-hemolytic streptococcus increased significantly. This could be due to an increase in the total number of patients who had varicella, an over-all increase in the rate of complications, or a change in the prevalence or the nature of group-A ß-hemolytic streptococcus. Although the number of hospital admissions for complications related to varicella increased in 1994, the ratio of admissions for musculoskeletal complications to admissions for complications related to varicella also increased significantly. As other authors have reported an increase in the number of musculoskeletal infections, most of which were due to group-A ß-hemolytic streptococcus, an actual change in the prevalence or the nature of this invasive organism is suspected3,21,29. A live attenuated varicella vaccine was recently approved by the United States Food and Drug Administration for children between the ages of twelve and eighteen months6. The vaccine was reported to be at least 70 per cent effective in preventing clinical varicella infection after exposure6. Moreover, when vaccinated patients do contract varicella, the clinical course appears to be milder, with fewer lesions6. The vaccine may decrease the prevalence of musculoskeletal infections. In the current series, there was no clear relationship between the number of varicella lesions and the development of musculoskeletal complications. Furthermore, infants younger than the age for immunization may still contract varicella. Four of the twenty-six children were eighteen months or younger. Additional studies of the prevalence and the nature of musculoskeletal complications in vaccinated children are warranted. In summary, this study provides additional evidence that varicella can result in life and limb-threatening complications in otherwise healthy children. It also documents group-A ß-hemolytic streptococcus as the predominant cause of musculoskeletal complications of varicella during an eleven-year period. The physician should have a high level of suspicion for musculoskeletal infection when examining children with varicella who have localized warmth and erythema, swelling, or pain or who refuse to bear weight. Prompt evaluation, appropriate operative intervention, and adjunctive systemic antibiotic therapy may prevent the spread of infection and the loss of life or limb.
NOTE: The authors thank Barbara Brand for assistance in the preparation of this manuscript. The authors also thank Christy Farnsworth and Tricia Silva for their technical contributions.
*No benefits in any form have been received or will be received from a commercial party related directly or indirectly to the subject of this article. No funds were received in support of this study.
  CiteULike  Connotea  Del.icio.us  Facebook  Technorati  Twitter What's this?
This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction
| home | CME | help | feedback | subscriptions | archive | site index |
|
Stanford University Libraries' HighWire Press (TM) assists in the publication of JBJS Online. |