Microbiology Case Study: 6 Year Old Girl with Headache

A 6 year-old girl with a history of posterior fossa ependymoma presented with a one month history of fever, headaches, vomiting and more recently, neck stiffness. Additional history includes remote tumor resection followed by radiation and chemotherapy resulting in remission, with a residual ventriculoperitoneal shunt (VPS). Her parents reported she was in good health until approximately 1 month prior to presentation and is up to date on her immunizations. She was previously seen by her primary care physician for her symptoms and treated her with amoxicillin for suspected strep throat. Upon admission, she received supportive therapy for her symptoms after she was found to have tumor recurrence on imaging. The patient was scheduled for resection approximately two weeks after discharge and on post-operative day two she developed fever, vomiting and neck stiffness again. At this time, blood cultures were drawn and a lumbar puncture (LP) was performed. Cerebrospinal fluid (CSF) from both the LP and VPS submitted for fluid analysis (Table 1) and culture.

Table 1: Cerebrospinal Fluid Analysis

Spinal Fluid LP VPS
Appearance Clear Clear
Nucleated cells 1075 cells/μL 628 cells/μL
RBC 150 cells/μL 35 cells/μL
Polys 94% 87%
Lymphs 2% 6%
Mono/Macrophage 4% 7%
Glucose 68 mg/dL 13 mg/dL
Protein 69 mg/dL 164 mg/dL
haem1
Figure 1. Gram stain of the pathogen isolated from aerobic blood culture, showing gram-negative coccobacilli, sometimes in pairs. The same organism was seen on the patient’s CSF Gram stain.
haem2
Figure 2. Aerobic blood culture on (A) 5% sheep blood agar plate (BAP), showing no growth and on (B) chocolate agar plate (CAP), showing round, smooth, opaque grey-yellow colonies.

 

Culture results:

The CSF Gram stain showed rare, paired, Gram-negative diplococci, which could raise suspicion for Neisseria meningitidis, however the typical flattened sides of adjacent bacteria were not observed. Rather, the morphology was more consistent with Gram-negative coccobaccilli, which is better demonstrated on Gram stain of the blood culture (Figure 1). Culture of both the CSF and blood specimens grew fairly large, smooth, round, opaque grey-yellow colonies on CAP, however showed no growth on BAP (Figure 2), suggesting a fastidious organism requiring growth factors. The colonies were both catalase and oxidase positive. The organism was identified as Haemophilus influenzae by MALDI-TOF MS (matrix-assisted laser desorption/ionizations time-of-flight mass spectrometry). This H. influenzae isolate was non-typeable by slide agglutination serotyping performed at the state public health laboratory.

Discussion:

H. influenzae are small, pleomorphic, gram-negative rods or coccobacilli that are non-motile. They are facultative anaerobes that grow best between 35-37°C with 5% CO2. H. influenzae is a fastidious species, requiring hemin (X factor) and nicotinamide-adenine-dinucleotide (NAD/V factor) for growth, which are both available in chocolate agar, but not blood agar. On chocolate agar, the colonies are non-hemolytic, typically large, smooth, round and convex with an opaque, colorless or grey hue. Encapsulated strains, including H. influenzae serotype b (Hib), appear mucoid and are typically small, grey colonies on CAP. Isolates are catalase and oxidase positive. H. influenzae displays the “satellite phenomenon” when grown near Staphylococcus aureus. This occurs when colonies of S. aureus lyse nearby red blood cells releasing hemin and NAD in the media. The presence of extracellular hemin and NAD allow colonies of H. influenzae to grow in the immediate vicinity of S. aureus.

H. influenzae is widely distributed in humans, colonizing the nose and throat and is spread from person-to-person via direct contact or respiratory droplets. Severe infections, including pneumonia, bacteremia and meningitis, affect predominantly infants and children. The American Academy of Pediatrics recommends routine vaccination with the Hib conjugate vaccine for infants aged 2 through 6 months (2 or 3 doses, depending on vaccine product) followed by a booster dose at age 12 through 15 months. Hib is the only serotype preventable by vaccine. Prior to routine vaccination in the US, approximately 20,000 children under the age of 5 were infected with H. influenzae and 3-6% died each year.

References:

1. Ledeboer N, Doern G. 2015. Haemophilus, p 667-684. In Jorgensen J, Pfaller M, Carroll K, Funke G, Landry M, Richter S, Warnock D (ed), Manual of Clinical Microbiology, 11th Edition. ASM Press, Washington, DC.
2. http://www.cdc.gov/meningitis/lab-manual/chpt09-id-characterization-hi.html
3. http://www.cdc.gov/hi-disease/index.html

 

-Petra Rahaman, MD is a 4th year Anatomic and Clinical Pathology resident at UT Southwestern Medical Center.

-Erin McElvania TeKippe, Ph.D., D(ABMM), is the Director of Clinical Microbiology at Children’s Medical Center in Dallas Texas and an Assistant Professor of Pathology and Pediatrics at University of Texas Southwestern Medical Center.

Microbiology Case Study: Immunocompromised Boy with Skin Nodules

An elementary school aged boy with a history of pre-B cell acute lymphocytic leukemia with a failed bone marrow transplant was transferred to a regional children’s hospital for leukodepletion and participation in an experimental clinical trial. At that time, his CBC was significant for 10% polymorphonuclear cells and 50% blasts. He was subsequently transferred to the ICU in respiratory failure and developed papulonecrotic lesions on his face, trunk, and bilateral legs. Prior to this, he was pancytopenic with no blasts present with cell counts of 100 WBC, hemoglobin 8.3 and 37,000 platelets. His Fungitell assay, which detects (1-3)-β-d-glucan, was positive.

Routine blood culture, fungal culture from the endotracheal tube, and fungal culture from the skin lesion biopsy specimens all had fungal elements on KOH stain. Young growth of a whitish, fluffy mold was present on all cultures within two days. Histopathology on the punch biopsy of a skin lesion on the thigh showed septate hyphae within the dermis, epidermis, and invading the vasculature that was particularly apparent with GMS stain (Figure 1a and 1b). Within a few days, the fungal cultures showed septate hyphae with microconidia using lactophenol cotton blue tape preparation, and shortly thereafter the mold developed into macroconidia with multiple septations taking on canoe-like forms (Figure 2). The white, cotton-like colonies developed a pink tinge (Figure 3). These characteristics allowed for the identification of the growth as Fusarium sp.

Septate hyphae on GMS stained section of the skin punch biopsy.
Septate hyphae on GMS stained section of the skin punch biopsy.
Septate hyphae on GMS stained section of the skin punch biopsy.
Septate hyphae on GMS stained section of the skin punch biopsy.
Microscopic identification of Fusarium by lactophenol cotton blue stain.
Microscopic identification of Fusarium by lactophenol cotton blue stain.
Colony of Fusarium growing on inhibitory mold agar (IMA).
Colony of Fusarium growing on inhibitory mold agar (IMA).

Fusarium is an opportunistic hyaline mold with infection most commonly seen in immunocompromised hosts. It can cause keratitis through contamination of contact lenses, penetration due to trauma, or use of immunosuppressive steroid ophthalmic solution. It is increasingly becoming the cause of disseminated infection in neutropenic hosts with a broader spectrum of disease, which includes: skin lesions, fungemia, rhinocerebral involvement and pneumonia. In these cases, without an immune system to fight the infection, mortality is high. Inhalation of airborne conidia, ingestion from water sources or access through mucosal membranes are all potential points of entry.

The colony growth on plated fungal media is rapid, usually maturing within four days. On microscopic examination, Fusarium hyphae are septate, approximately 3-6 microns wide with acute angle branching. Microconidia are small, oval-shaped, and no larger than 4 x 8 microns in size. These can look like Acremonium sp. Macroconidia are canoe- or sickle-shaped with the largest dimension being about 80 microns in length, exhibiting 3-5 septatations.

 

Jodi Music, MD, is an AP/CP resident at UT Southwestern Medical Center.

-Erin McElvania TeKippe, Ph.D., D(ABMM), is the Director of Clinical Microbiology at Children’s Medical Center in Dallas Texas and an Assistant Professor of Pathology and Pediatrics at University of Texas Southwestern Medical Center.

Microbiology Case Study: An 8-Week-Old Female with Pallor, Vomiting, Fever, and Blue Feet

Clinical History

An 8 week old female was brought to an outside hospital due to pallor, decreased eating over several days, vomiting, fever, and blue color in the feet. The patient had received her two month vaccinations the day prior to presentation. Her past medical history was significant for being born at 32 weeks gestation, followed by an uneventful 4 week NICU stay. At the outside hospital the patient was in respiratory distress, tachycardic, with pallor. She was intubated and transferred to our institution due to concern for an ALTE (apparent life-threatening event). Blood and CSF specimens were drawn. Upon presentation, the patient had a white blood cell count of 19,600/mm3. Her CSF had 63 nucleated cells/mm3 (30% neutrophils, 49% lymphocytes, 11% monocytes), glucose of 23 mg/dL and protein of 212 mg/dL. Blood and CSF cultures were performed with the following results:

Gram stain of cerebral spinal fluid (CSF) specimen showing Gram postive cocci in singles and pairs
Gram stain of cerebral spinal fluid (CSF) specimen showing Gram positive cocci in singles and pairs
Gram stain of positive blood culture broth showing Gram positive cocci in chains
Gram stain of positive blood culture broth showing Gram positive cocci in chains
Small zones of beta hemolysis around colonies growing on 5% sheep blood agar
Small zones of beta hemolysis around colonies growing on 5% sheep blood agar

Laboratory Identification

CSF Gram stain was read as many Gram-variable coccobacilli (Image 1) and blood culture Gram stain was resulted as Gram-positive cocci in pairs and chains (Image 2). The Nanosphere Verigene Gram-positive blood culture molecular assay was performed on the positive blood culture broth immediately following Gram staining. CSF and positive blood culture specimens were plated to 5% sheep blood, chocolate, MacConkey, and CNA agars. In >24 hours colonies from both specimens grew a single organism on sheep blood, chocolate, and CNA agars. On blood agar, colonies exhibited a soft zone of b-heomlysis (Image 3). Colony Gram stains showed Gram-positive cocci in chains, catalase testing was negative, and the organism typed in Lancefield antigen group B. Verigene identified the organism directly from the positive blood culture broth as Streptococcus agalactiae (aka Group B Streptococcus or GBS) and MALDI-TOF mass spectrometry confirmed the identification of S. agalactiae.

Discussion

Laboratory Considerations

As our CSF specimen demonstrated, streptococci can be difficult to interpret from specimen Gram stains. Organisms are often are pleomorphic in size and shape and they have a tendency to stain Gram-variable. This lead to the report of Gram-variable coccobacilli on our patient’s CSF culture.

S. agalactiae produces a soft zone of b-heomlysis on sheep blood agar. Unlike Streptococcus pyogenes (aka Group A Streptococcus) which produces a wide zone of b-hemolysis, soft b-hemolysis can often be very subtle, especially with young growth. Soft zones b-hemolysis can best be seen by holding plates up to a light source or my moving a colony out of the way to observe if hemolysis is present underneath. The pattern of S. agalactiae b-hemolysis is very similar to that produced by Listeria monocytogenes. Streptococcal isolates that type as Lancefield Group B, but produce large zones of b-hemolysis can create confusion and are most likely not S. agalactiae, but S. porcinus or S. pseudoporcinus.

Clinical Significance

S. agalactiae is the cause of significant neonatal disease. Early-onset infection presents as systemic infection, respiratory distress, apnea, shock, and pneumonia within the first 24 hours of life (range, 0–6 days). Meningitis is less common in early-onsetS. agalactiae infections, found in just 5-10% of cases. Late-onset disease presents at 3 to 4 weeks of age (range, 7–89 days) as meningitis and/or sepsis with other focal infection. Approximately 50% of survivors of early- or late-onset meningitis have long-term neurologic sequelae.

S. agalactiae colonizes the urogenital or gastrointestinal track of 10-30% of pregnant women. Being born to a S. agalactiae colonized mother is the most significant risk factor for development of disease in neonates. For this reason, women are screened for “Group B Strep” colonization between 35 and 37 weeks of pregnancy. Women found to be colonized receive prophylactic antibiotics immediately prior to delivery to prevent transmission to the child. Since implementation of these practices in 1996, there has been a substantial decline in early-onset S. agalactiae infections. Interestingly, these measures have not affected the incidence of late-onset disease.

Treatment

Until recently S. agalactiae was considered universally susceptible to penicillin. There have now been a few reports of S. agalactiae isolates with increased penicillin MICs due to mutations in the penicillin binding protein Pbp2x. The detection of these isolates is still extremely rare, so much so that S. agalactiae susceptibility testing for penicillin and other b-lactams is not considered necessary at this time. Penicillin and its derivatives are the preferred treatment option.

Patient follow-up

Our patient had a complicated course of late-onset Group B Streptococcal meningitis including multifocal cerebral infarctions and seizures. She was treated with a 28 day course of ampicillin. Blood cultures taken 24 hours after the start of antibiotics were negative and her CSF culture was negative when rechecked 5 days after her presentation. Due to her complicated course, the patient was hospitalized for 4 weeks. Follow-up appointments have shown the patient’s MRI is nearly normal 8 weeks post infection and the patient is doing well, although she is still followed by neurology to assess for long term sequelae.

Our patient’s mother had an unknown Group B Strep carrier status at the time of her birth, as she was born at 32 weeks, which is before routine screening occurs for pregnant women. Either way, the mother’s status would not have affected the patient’s risk for late-onset Group B Streptococcal infection.

References:

  1. Red Book 2015
  2. Journal of Clinical Microbiology, 10th edition

-Erin McElvania TeKippe, Ph.D., D(ABMM), is the Director of Clinical Microbiology at Children’s Medical Center in Dallas Texas and an Assistant Professor of Pathology and Pediatrics at University of Texas Southwestern Medical Center.

Microbiology Case Study: an 11 Month Old with a 3 Day History of Gastroenteritis and High Fever

Clinical History:

An 11 month old female with no significant past medical history was admitted with a fever of 104 degrees Fahrenheit, nausea and vomiting for 3 days (now resolved), watery diarrhea 4-5 times/day (resolved), and a new onset of acute pharyngitis/bilateral cervical adenitis. ER staff was concerned for a bacterial superinfection. She appeared sick with pale skin, but vital signs were stable, and labs were unremarkable except for an elevated CRP (15.7) and an absolute monocytosis (though no elevation in total WBCs). Exam showed a hyperemic pharynx without exudates, and no lymph nodes larger than 1 cm. A CT neck shows bilateral cervical adenitis, left greater than right, with some suggestion of necrotic nodes, as well as a likely left 3rd or 4th branchial cleft cyst. Blood cultures were drawn, and they turned up positive in a matter of hours, with the gram stain and plate morphology seen below:

Gram stain of positive blood culture broth showing Gram positive cocci in chains
Gram stain of positive blood culture broth showing Gram positive cocci in chains
Large zones of beta-hemolysis around colonies growing on 5% sheep blood agar
Large zones of beta-hemolysis around colonies growing on 5% sheep blood agar

Laboratory Identification:

Gram positive cocci in chains were seen, with small, glossy, gray-white, translucent colonies on blood agar having a wide zone of surrounding beta hemolysis. Catalase testing was negative, PYR testing was positive, and latex agglutination testing for Lancefield antigens was positive for Group A. MALDI-TOF confirmed the presumptive identification of Streptococcus pyogenes.

Discussion:

S. pyogenes (aka Group A Streptococcus [GAS]) is a ubiquitous gram positive cocci that causes a wide range of disease in humans. It is the leading cause of acute pharyngitis, particularly in children aged 5-15, although 15-25% of school aged children are asymptomatically colonized. S. pyogenes can also cause cellulitis, impetigo, necrotizing fasciitis, scarlet fever, toxic shock syndrome, otitis media, osteomyelitis, pneumonia, or even rarely meningitis/brain abscess. It additionally causes several serious post-infectious sequelae, particularly in untreated cases, including acute rheumatic fever with potential rheumatic heart disease, and poststreptococcal glomerulonephritis.

S. pyogenes has numerous virulence factors, most importantly the M protein, of which there are nearly 80 variants. M protein binds fibrinogen, inhibits complement binding, and prevents phagocytosis. As it is the major antigenic target of antibodies formed following infection, immunity is conferred only to the infecting strain, and none of the others, which complicates vaccine development efforts. Streptolysin O is responsible for the organism’s b-hemolysis on blood agar, and is also immunogenic, causing elevated Anti-Streptolysin O antibodies (ASO), which can be useful in diagnosing a recent infection, rheumatic fever, or poststreptococcal glomerulonephritis.

GAS pharyngitis is uncommon in children less than 3 years old, as is the bacteremia seen in the patient, which was presumably secondary to the severity of the throat infection. A later CT scan showed an enlarging abscess in the left lateral neck, corresponding to the earlier supposed branchial cleft cyst, despite IV clindamycin which was begun in the ER. Ceftriaxone was added, and later changed to Piperacillin/Tazobactam to complete a seven day course, though future blood cultures were negative. The additional antimicrobial coverage was due to the concern for other infectious microorganisms, as S. pyogenes is universally susceptible to penicillin.

Interestingly, this patient also developed a severe absolute neutropenia, with her ANC dropping from 7,100 at admission to 300 two days later, and then to 60 after two further days. The hematology/oncology service was consulted, and they determined that this likely represented a reaction to the infection rather than a more sinister bone marrow pathology. Several more days into therapy her ANC did begin to recover.

References:

  1. Henningham A, Barnett TC, Maamary PG, Walker MJ. 2012. Pathogenesis of group A streptococcal infections. Discovery medicine 13:329-342.
  2. Cunningham MW. 2000. Pathogenesis of group A streptococcal infections. Clinical microbiology reviews 13:470-511.
  3. Red Book 2015
  4. Journal of Clinical Microbiology, 10th editionCunningham MW. 2000. Pathogenesis of group A streptococcal infections. Clinical microbiology reviews 13:470-511.

-Paul Yell, M.D. is a 2nd year anatomic and clinical pathology resident at the University of Texas Southwestern Medical Center.

-Erin McElvania TeKippe, Ph.D., D(ABMM), is the Director of Clinical Microbiology at Children’s Medical Center in Dallas Texas and an Assistant Professor of Pathology and Pediatrics at University of Texas Southwestern Medical Center.

Pediatric Labs are Different

Why do I say pediatric labs are different? A blood test is a blood test, right? Of course it is, however there are a lot of aspects of operating a pediatric lab that really are different from the way operations run in a lab whose clientele is mainly adults.

Most people can think of the most obvious difference: age-related reference intervals. The concentration of various biomarkers in the body changes as an infant grows and develops into an adult. In some cases, there’s not a lot of change in the biomarker. A blood pH is pretty constant over the course of a person’s life, as are the electrolytes. The reference intervals for these may be broader in infancy, but in general they don’t change a lot. However some biomarkers change so drastically that normal levels in childhood would be considered pathological in an adult. Without reference intervals tied to age or developmental state, these tests are not able to be properly interpreted. Alkaline phosphatase during bone growth and steroid hormones during puberty are two good examples of this.

The things that people tend not to think about that cause a pediatric lab to be different are predominately all centered on issues with sample volume. Especially in infancy, children have limited blood volume for testing, and this fact affects nearly every operation in a pediatric laboratory.

Front end processing of samples is affected, as often more than 60% of the tubes arriving in the lab are microtubes, which hold 1 mL or less. These tubes may not be able to hold a barcode label, nor fit on an instrument robotic system. The sample must be aliquotted into tubes that will fit a system, or hand fed into an instrument, and often hand-programmed into the instrument. More manual steps results in more opportunities for human error. In addition, the amount of sample used by any given instrument for testing must be considered. That sample volume includes the actual volume necessary for analysis, plus the volume necessary to run HIL (hemolysis, icterus, and lipemia) indices and the instrument dead volume (the volume below which an instrument cannot pipet the sample). Instruments with large dead volumes will not be found in pediatric labs.

Small sample volumes also affect the ability of the lab to add-on tests to samples already in the lab, or rerun samples later to check results. There may be insufficient volume to run more tests or rerun tests, or the sample may have evaporated and be unacceptable for running additional analyses. A study done using a 5 mL serum sample and a 0.1 mL serum sample and allowing both to sit open to the air for 4 hours showed that the 5 mL serum sample had less than 10% difference between original test values and those run 4 hours later. The 0.1 mL sample had more than 50% difference in its values. In addition, with just enough sample to run a test one time, if a dilution needs to be made, it may not be possible.

Lastly, there are a few test menu differences in a pediatric laboratory. Tests commonly found in pediatric labs, but essentially never in predominately adult labs include things like testing for inborn errors of metabolism and sweat chloride testing for Cystic Fibrosis. Conversely, common tests in an adult lab that are not performed in a pediatric lab include serum protein electrophoresis, PSA and other tumor markers. Some tests are performed in both labs but for different indications. For example, AFP is used as a tumor marker for hepatocellular carcinoma in pediatrics instead of as a maternal prenatal screening tool like in an adult lab.

These are some of the ways in which pediatric lab medicine differs from adult lab medicine, and offers unique challenges in day to day operations.

 

-Patti Jones PhD, DABCC, FACB, is the Clinical Director of the Chemistry and Metabolic Disease Laboratories at Children’s Medical Center in Dallas, TX and a Professor of Pathology at University of Texas Southwestern Medical Center in Dallas.

Pediatric Reference Intervals

All laboratory professionals are aware of the importance of reference intervals (RI). Without appropriate reference intervals, a test result is just a number. A numeric result for any given analyte cannot be used to diagnose or monitor treatment of disease unless there is an accompanying reference interval indicating what amount of that analyte should normally be present. In pediatrics, that’s even more important because as an infant develops through childhood and into adulthood, his or her biochemistry changes, adapts and develops with him or her. Using an adult RI to interpret a test result from an infant or child is likely to result in misinterpretation of the test, including misdiagnosis of disease states. A good example would be using adult alkaline phosphatase RI to interpret the results of a teenage boy’s test during accelerated bone growth in puberty.  His result will look pathological if interpreted using adult RI. Pediatric reference intervals (PRI) are age-related and often also gender-related intervals that must be used to interpret testing in the pediatric population.

Establishing reference intervals is problematic at the best of times because of the need to use 120 healthy individuals to establish them. In the pediatric population, especially in infancy, obtaining 120 healthy infants at each necessary time interval can be a daunting task. There are references available that present methods which allow the establishment of RI with smaller sample numbers (1,2). Also the CLSI document (2) allows the “validation” and “transfer” of current or literature RI, rather than complete “establishment” of RI in some cases.  Validating a current reference interval can be done with as few as 20 samples in a correlation study. “Transferring” a reference interval also involves using a correlation study and bias evaluation to adapt or adjust a current RI for use with a new assay. Transferring can also be performed with 20 – 60 patient samples.

These techniques especially come in handy with the hardest PRI to establish, the hormones during puberty. During this time, the RI are not really related to the child’s age, but related to the child’s phase of development, or Tanner Stage. To establish a PRI for these hormones, the healthy child donating the blood sample must also have his/her Tanner Stage determined, usually by an Endocrinologist.

Another consideration when dealing with PRI is that although all pediatric institutions use PRI, not all PRI are the same, even when the same instrument is used. An informal poll of 9 pediatric institutions using the same instrument resulted in 9 different PRI for common analytes such as electrolytes. There is a need to harmonize PRI, as we harmonize test results, in order to allow non-pediatric institutions a set of PRI that they can use also.

  1. Horn PS, Pesce AJ. Reference Intervals: a User’s Guide. AACC Press, Washington DC, 2005
  2. Defining, Establishing, and Verifying Reference Intervals in the Clinical Laboratory: Approved Guideline – Third ed. CLSI C28-A3

 

???????????????????????????????????????????????????????????????????????????????????

-Patti Jones PhD, DABCC, FACB, is the Clinical Director of the Chemistry and Metabolic Disease Laboratories at Children’s Medical Center in Dallas, TX and a Professor of Pathology at University of Texas Southwestern Medical Center in Dallas.