Unresponsive, toddler presents to PICU with fever

James H. Brien

A previously healthy 2-year-old girl presented with a 1-day history of high fever of 103°F at home, followed by some episodes of diarrhea. In less than 24 hours, she was found to be unresponsive and was taken by ambulance to the local ED.

From there, she was transferred to the children’s hospital pediatric ICU, where she was intubated due to poor respiratory effort. A full sepsis work-up was performed, including cultures of blood, urine and spinal fluid, and she was empirically given vancomycin and ceftriaxone.

Examination:

Vital signs on admission revealed a temperature of 105.8°F, a pulse of 177, a prolonged capillary refill with a petechial rash (Figure 1) and mottled skin. She was in respiratory failure. There were no other pertinent findings on exam, but the rash soon progressed into large areas of purpura (Figure 2).

Figure 1. Petechial rash. Source: James H. Brien, DO.

Figure 2. Purpura. Source: James H. Brien, DO.

Additional history revealed that she attends day care but had no known sick contacts there. However, her older brother at home had an influenza-like illness about 1 week earlier (he was not seen, and the illness had since resolved). At that same time, the patient also had a brief, mild upper respiratory infection (URI), but she was not acting sick. Her travel history is unremarkable, and the only animal contact was with the healthy family dog. Her immunizations are up to date through her 18-month visit, and she takes no routine medication. Her diarrhea continued with some hematochezia, and her urine output decreased. She was admitted with altered mental status and is now sedated on the ventilator.

Her admitting lab results:

1. Hematology: CBC is 27.1, WBC is a predominance of granulocytes, and there is evidence of disseminated intravascular coagulation (DIC) on coagulation studies.
2. Chemistries: Testing shows mildly elevated creatinine and liver enzymes.
3. The urine was concentrated but otherwise normal, as was the CSF analysis.

Her chest radiograph was normal, showing only the endotracheal tube placement.

Her blood culture was reported positive in less than 24 hours, with chains of gram-positive cocci seen on Gram stain (Figure 3).

Figure 3. Gram-positive cocci in chains. Source: James W. Bass, MD.

Summary:

1. The patient is a previously healthy, fully immunized 2-year-old with a 1-day history of high fever and some bloody diarrhea, altered mental status with rapid deterioration, requiring PICU admission and intubation due to septic shock.
2. She progressed to acute renal insufficiency, transaminitis, respiratory failure and DIC.
3. Her illness was preceded by mild URI symptoms a week earlier.
4. Blood culture grew gram-positive cocci in chains in less than 24 hours.

What’s your diagnosis (cause for sepsis)?

A. Neisseria meningitidis
B. Staphylococcus aureus
C. Haemophilus influenzae type b
D. Streptococcus pyogenes

Answer and discussion:

This is a case of Streptococcus pyogenes sepsis (choice D), with multiorgan involvement, consistent with toxic shock syndrome and a blood culture rapidly growing characteristic gram-positive cocci in chains. In the classic presentation of sudden sepsis with rapid progression to purpura fulminans, one is tempted to think meningococcemia; however there is no reliable way to distinguish the causative organism based on the clinical presentation, as all the organisms listed can present in this fashion. Additionally, N. meningitidis would be showing gram-negative diplococci (Figure 4). One can virtually rule out H. influenzae type b (choice C) by the fact that the patient’s immunizations are documented as up to date, implying that she has had all recommended Hib doses. While one may rarely see invasive non-type b H. influenzae with sepsis or meningitis, type b is almost unheard of in the fully immunized child. Also, the blood culture would be growing gram-negative pleomorphic rods, as shown in Figure 5. However, in the unimmunized child, sepsis with purpura fulminans can be caused by Hib, as seen in Figure 6. In either case, one would not want to bet the child’s life on the immunization record and therefore should empirically treat for Hib as well until culture results are available. Lastly, Staphylococcus aureus, originally described as the toxic shock organism, would be gram-positive cocci in clusters. Always bear in mind that Gram stains can also be misleading due to technique or interpretation. Therefore, therapy or changes should always be based on culture and sensitivity results.

Figure 4. Gram-negative diplococci. Source: James W. Bass, MD.

Figure 5. Gram-negative pleomorphic rods. Source: James W. Bass, MD.

Figure 6. Purpura fulminans due to H. influenzae type b. Source: James W. Bass, MD.

If you practice medicine long enough, you will see various infectious diseases appear to wax and wane through the decades, as has been the case with scarlet fever and invasive group A strep (GAS) infections. Not much is known about the disease before the 19th century except that of a febrile disease with a red rash. Through most of the 1800s, scarlet fever was a dreaded childhood disease, with significant mortality. The death rate was no doubt increased in part due to the popular practice of bloodletting. In any case, periodic spikes in severe, invasive scarlet fever (GAS disease) have been described for over 150 years. There appeared to be a dramatic decrease in severe scarlet fever around 1870, only to have a brief comeback in the early 20th century. During that time, scarlet fever was divided into three clinical types: simple, anginoid and malignant.

The simple variety was a sore throat with fever and a fine, papular red rash with Pastia lines and a “strawberry tongue,” typically lasting 5 to 7 days with generalized desquamation. The anginoid variety was the same but with higher fever, more severe sort throat with ulcerations and profound prostration. Death would occasionally occur as a result of “exhaustion” or aspiration pneumonia. The malignant variety was associated with chills and fever reaching 107°F (sepsis?), vomiting, diarrhea, seizures, bleeding from the membranes and skin (petechiae and purpura), nephritis and other organ dysfunction (toxic shock?). Death was common with the malignant scarlet fever.

By the mid-20th century, scarlet fever became a fairly benign disease again. Then again severe disease was increasingly seen in the last 2 decades of the 20th century. In more recent years, an outbreak of severe disease in England and Wales brought about the discovery of a new strain of GAS, called M1UK (Lynskey and colleagues). Shifts in genetics and virulence factors no doubt were taking place throughout the entire history of this disease. The purpose of this discussion is to make the reader aware of this constant shift in virulence of this organism that to some may still be frequently thought of as little more than a sore throat-causing nuisance. Many practitioners have been taught that scarlet fever is simply strep throat with a rash. In many cases, that is correct. However, like many diseases, scarlet fever is a descriptive term for this toxin-mediated infection, which has a wide range of severe manifestations. Severe scarlet fever may mean GAS sepsis or toxic shock syndrome, with purpura fulminans, or as the physicians of yesteryear called it, malignant scarlet fever.

All ID practitioners know that rapidly progressive septic shock with purpura in an otherwise healthy child is not the sole domain of N. meningitidis, and never has been. The purpose with this case is to draw attention to this fact for those who may only see one of these children come to their office during their career. Any sepsis protocol that is followed will likely include an antibiotic effective against GAS. However, with rash, multiorgan impairment or any suggestion of a toxin-mediated infection, many experts recommend adding clindamycin to the empiric regimen until the cause is known, and improvement over several days occurs. This is done in the hopes of down-regulating the toxin production from the 50S ribosome of the organism, which sets into motion the cascade of steps that is responsible for many of the adverse effects. For this reason, the empiric addition of clindamycin has been included in many septic shock protocols where a rash-producing toxin appears to be present, as was done in this case.

The patient in this case survived but lost her left leg below the knee.

Columnist comments:

If you have a good case and you would like to see featured in this column, just write me at jhbrine@aol.com, and I’ll help make it happen.

For more information:

Brien is a member of the Infectious Diseases in Children and Infectious Disease News Editorial Boards, and an adjunct professor of pediatric infectious diseases at McLane Children's Hospital, Baylor Scott & White Health, in Temple, Texas. He can be reached at jhbrien@aol.com.

References:

Lynskey NN, et al. Lancet Infect Dis. 2019;doi:10.1016/S1473-3099(19)30446-3.

Stevens AA. A Manual of the Practice of Medicine. Saunders, 1934: 356-361.