Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Clin Infect Dis. Author manuscript; available in PMC 2010 December 15.
Published in final edited form as:
PMCID: PMC2787872

Staphylococcal Toxic Shock Syndrome Erythroderma is Associated with Superantigenicity and Hypersensitivity


Staphylococcal toxic shock syndrome (TSS) has rarely been reported without rash and desquamation. This study describes a patient who met all criteria for TSS except erythroderma and desquamation. The associated staphylococcal superantigen was enterotoxin B. We demonstrate erythroderma depends on pre-existing T cell hypersensitivity amplified by superantigenicity.

Keywords: Staphylococcus aureus, toxic shock syndrome, erythroderma, superantigen

Staphylococcal toxic shock syndrome (TSS) is defined by fever, hypotension, erythroderma, desquamation, and variable multiorgan components [1]. Menstrual TSS occurs primarily in women using tampons [2] and is associated with the superantigen TSS toxin-1 (TSST-1) [3]. Non-menstrual TSS occurs in males and females initiated by any type of infection [4]; cases are associated with TSST-1 and staphylococcal enterotoxins B (SEB) and C [3].

When TSS was identified, it was recognized that cases occur where one defining criterion is absent; these cases are defined as probable TSS [4]. However, little attention has been paid to cases where multiple criteria are absent. Parsonnet suggests these cases be identified as toxin-mediated disease [5].

We describe a TSS patient whose initial diagnosis was difficult because erythroderma and desquamation were absent. Once identified as variant illness, the patient responded to intravenous immunoglobulin (IVIG) and clindamycin. We demonstrate that erythroderma depends on delayed hypersensitivity amplified by superantigenicity.

Materials and Methods

The patient’s S. aureus was tested by PCR for genes for superantigens and Panton-Valentine leukocidin (PVL) [6], and by quantitative antibody assay for SEB after growth (Todd Hewitt broth; Becton, Dickinson and Company, Sparks, MD) [7]. Positive and negative control strains responded as expected in both tests. Patient serum was tested by ELISA for SEB antibodies prior to administration of IVIG.

Eight Dutch-belted rabbits were tested with 1 ug/0.1 ml highly purified SEB intradermally on their flanks, and were monitored for 48 hr for erythroderma. The same animals were then given a subcutaneous sensitizing dose of SEB (25 ug) in Freund’s incomplete adjuvant (Difco, Detroit, MI). After 2 weeks, the rabbits were injected intradermally with SEB (1 ug/0.1 ml), or SEB pre-mixed for 30 min with 10 ug soluble high-affinity variable region, β-chain T cell receptor (Vβ-TCR designated G5-8) capable of neutralizing superantigenicity [8]; animals were observed for erythroderma.

Case Report

A 16 year-old otherwise healthy teenager presented to urgent care with complaints of headache, deep breathing, and nausea for three days and increasing diffuse weakness for 24 hr. She denied fever, recent weight loss, vomiting, diarrhea, polyuria, or polydipsia. Past history revealed no illnesses, and she was not on medications. She began menses at age 11, described her cycles as “irregular”, and denied tampon use.

Her temperature was 95.5° F, blood pressure 119/80, pulse 117, respirations 28, with 100% oxygen saturation on room air. She was awake, in no acute distress, and cooperative. Her examination was notable for Kussmaul respirations and a 3–4 second capillary refill. No erythroderma was seen. The remainder of the exam was normal.

Initial laboratory data revealed metabolic acidosis (pH 6.77), with hyperglycemia glucose level, 524 mg/dL). She was diagnosed with diabetic ketoacidosis (DKA) and treated with IV fluids and insulin. Her total white cell count was 7.6 × 109/L with differential of 82% neutrophils, 8% lymphocytes, and 9% monocytes. Hemoglobin was 15.9 mg/dL, and platelets numbered 158,000. Urinalysis demonstrated glucose of 300 mg/dL, ketones of 10 mg/dL, 14 red blood cells, 1 white blood cell, few bacteria, and amorphous crystals. A foley catheter was inserted to monitor urine output.

She was transferred to the pediatric intensive care unit (PICU). Three hours later, her condition worsened. She became combative and confused. Her serum osmolarity was 314, and IV mannitol was administered with resolution of her mental status changes. Shortly thereafter, she had a fever of 38.5°C. No antibiotics were started, and no cultures obtained. After 18 hr in the PICU, she again became combative and confused. Mannitol and bicarbonate were given without effect. Central access was obtained with a femoral venous catheter, and she was intubated. Subsequently, she became hypotensive (blood pressure 77/39) and was started on a norepinephrine drip. After 9 hr, she developed fever to 40.0°C. Blood, endotracheal, and urine cultures were obtained, and IV cefotaxime and vancomycin were started. Computed tomography scan revealed no cerebral edema.

She developed thrombocytopenia (platelets 25,000/mm3), acute renal failure (creatinine 1.98 mg/dL), and persistent hypotension requiring phenylephrine, norepinephrine, and dopamine 24 hr after admission. Elevations of creatine kinase (833 mg/dL), lipase (371 U/L), amylase (738 U/L), and alanine transaminase (103 U/L) were noted. Her hemoglobin A1C level was 15.5%. Chest x-ray revealed no abnormalities. Due to concern about sepsis, caspofungin and meropenem were started; cefotaxime was discontinued. A serum toxicology screen and testing for serum salicylate and acetaminophen, were all negative.

On day 3 of hospitalization, an endotracheal tube sputum culture taken shortly after intubation grew methicillin-sensitive S. aureus (MSSA), a urine culture grew 10,000–50,000 MSSA, and the next day the initial blood culture grew MSSA. TSS was suspected despite the lack of erythroderma. Clindamycin and cefazolin were started, and IVIG was given (1gm/kg). Other antimicrobials were stopped. Further examination revealed no vaginal foreign objects, and repeat ECHO demonstrated no vegetations or regurgitation. She lacked mucosal hyperemia during hospitalization.

Within 6 hr of receiving IVIG and 9 hr of receiving clindamycin, she was weaned off phenylephrine; within 12 hr, the norepinephrine drip was being weaned. Three days after receiving IVIG and clindamycin, her hypotension resolved, she was extubated to room air, and her mental status returned to baseline. Blood cultures taken over 4 days after initial blood culture were negative. A total of 5 days of clindamycin and 2 weeks of cefazolin were given. A 0.5 cm blister was noted on her right thumb 6 days after admission, but no other desquamation was noted, including the 1–2 weeks specified in the TSS definition.

Superantigen Studies

Her S. aureus contained genes for SEB and SE-like G, K, L, and N but not PVL. The strain produced SEB (77 ug/ml) in vitro; ELISA demonstrated low SEB antibodies titers (≤1:40, compared to 1:640 for IVIG). Approximately 10% of S. aureus strains produce SEB, and 90% of 15–20 year-old females should have titers >1:40 to SEB [9].

Eight rabbits were skin-tested with 1 ug/0.1ml of SEB; none showed erythroderma. The same animals were sensitized to SEB for 2 weeks and then re-tested. Four rabbits received SEB and showed erythroderma (diameters 11 ± 1.3 cm; example in figure 1). The remaining four rabbits received SEB pre-mixed with 10 ug soluble Vβ-TCR. These animals showed minor erythroderma (0.2 ± 0.3 cm; example in figure 1) (p<<0.001 compared to treatment with SEB by Student’s t test).

Figure 1
Skin test reactivity of a rabbit to SEB, and neutralization by soluble high-affinity Vβ-TCR (G5-8). Rabbits were immunized subcutaneously in the nape of the neck for two weeks with SEB (25 ug) emulsified in incomplete adjuvant and then challenged ...


TSS includes the characteristic erythroderma, fever, hypotension, desquamation, and multisystem organ involvement [1, 2]. A probable diagnosis can be made with one criterion absent. Our patient lacked erythroderma and desquamation, and thus did not meet the criteria for TSS or probable TSS. A small blister on her right hand was not felt to meet criteria for desquamation. However, the patient met all other criteria for TSS. No pathogens other than S. aureus were grown, and the patient’s isolate produced SEB, a superantigen associated with non-menstrual TSS [3]. Finally, the patient’s serum prior to IVIG therapy contained low SEB antibodies, so the patient was serosusceptible.

Because the patient did not have erythroderma, her clinical presentation was initially puzzling. Without initial fever, it is unclear whether S. aureus triggered DKA, or whether she became superinfected while in DKA. Her symptoms progressed to septic shock, but without the characteristic rash, it was difficult to implicate staphylococcal TSS; although it had been considered. With positive cultures, however, the diagnosis became evident.

The dramatic improvement after IVIG and clindamycin suggests a response to these treatments. Although we cannot know with certainty if clinical improvement was related to the treatments, anecdotal case reports and in vitro studies suggest IVIG and clindamycin may be effective therapies for TSS. Further data on their efficacy is required before they can be accepted as standard of care.

There are few full case reports of staphylococcal TSS without rash. Although early epidemiologic reports note a few cases lacking rashes [4], Van Lierde et al. [10] and Matsuda et al. [11] published full clinical descriptions of individual cases. Kamel et al. concluded that the absence of rash in three TSS patients, who underwent chemotherapy for multiple myeloma, reflected T cell deficiency [12].

An alternative explanation for the absence of erythroderma is that it results from delayed hypersensitivity amplified by superantigenicity. Some women describe partial episodes of menstrual TSS symptoms preceding definite TSS. In those prior episodes, erythroderma and desquamation may be absent, which is consistent with the need for hypersensitivity. The isolation of SEB from our patient’s S. aureus strain indicates the illness did not originate from vaginal colonization. It is more likely that this is the patient’s first encounter at a non-vaginal site with an SEB+ organism.

The rabbit experiments demonstrate erythroderma is associated with delayed hypersensitivity; the skin tests peaked in 24–48 hr. The studies with co-administration of SEB-neutralizing, high-affinity Vβ-TCRs (G5-8) demonstrate that superantigenicity is also required [8].

The present case highlights that TSS may present without erythroderma and desquamation, with cases more frequent than presently recognized. The case also highlights the need for further studies on whether IVIG and clindamycin decrease morbidity and mortality.


This work was supported by USPHS research grants U54-AI57153 (Great Lakes Regional Center of Excellence in Biodefense and Emerging Infectious Diseases where both PMS and DMK are members) and R01-AI064611 from the National Institute of Allergy and Infectious Diseases.


Conflict of Interest

None of the authors have conflicts of interest to report.


1. Syndrome TSS. In: Red Book: Report of the Committee on Infectious Diseases. Pickering LK, Baker CJ, Long SS, McMillan JA, editors. Vol. 27. Elk Grove Village, IL: American Academy of Pediatrics; 2006. p. 662.
2. Shands KN, Schmid GP, Dan BB, et al. Toxic-shock syndrome in menstruating women: association with tampon use and Staphylococcus aureus and clinical features in 52 cases. N Engl J Med. 1980;303:1436–1442. [PubMed]
3. Schlievert PM, Tripp TJ, Peterson ML. Reemergence of staphylococcal toxic shock syndrome in Minneapolis-St. Paul, Minnesota, during the 2000–2003 surveillance period. J Clin Microbiol. 2004;42:2875–2876. [PMC free article] [PubMed]
4. Reingold AL, Hargrett NT, Dan BB, Shands KN, Strickland BY, Broome CV. Nonmenstrual toxic shock syndrome: a review of 130 cases. Ann Intern Med. 1982;96:871–874. [PubMed]
5. Parsonnet J. Case definition of staphylococcal TSS: a proposed revision incorporating laboratory findings. International Congress and Symposium Series. 1998;229:15.
6. Schlievert PM, Case LC. Molecular analysis of staphylococcal superantigens. Methods Mol Biol. 2007;391:113–126. [PubMed]
7. Schlievert PM. Immunochemical assays for toxic shock syndrome toxin-1. Methods Enzymol. 1988;165:339–344. [PubMed]
8. Buonpane RA, Churchill HR, Moza B, et al. Neutralization of staphylococcal enterotoxin B by soluble, high-affinity receptor antagonists. Nat Med. 2007;13:725–729. [PubMed]
9. Schroder E, Kunstmann G, Hasbach H, Pulverer G. Prevalence of serum antibodies to toxic-shock-syndrome-toxin-1 and to staphylococcal enterotoxins A, B and C in West-Germany. Zentralbl Bakteriol Mikrobiol Hyg [A] 1988;270:110–114. [PubMed]
10. Van Lierde S, van Leeuwen WJ, Ceuppens J, Cornette L, Goubau P, Van Eldere J. Toxic shock syndrome without rash in a young child: link with syndrome of hemorrhagic shock and encephalopathy? J Pediatr. 1997;131:130–134. [PubMed]
11. Matsuda Y, Kato H, Yamada R, et al. Early and definitive diagnosis of toxic shock syndrome by detection of marked expansion of T-cell-receptor VBeta2-positive T cells. Emerg Infect Dis. 2003;9:387–389. [PMC free article] [PubMed]
12. Kamel NS, Banks MC, Dosik A, Ursea D, Yarilina AA, Posnett DN. Lack of muco-cutaneous signs of toxic shock syndrome when T cells are absent: S. aureus shock in immunodeficient adults with multiple myeloma. Clin Exp Immunol. 2002;128:131–139. [PubMed]