May 2006

A 2-year-old girl was admitted to the pediatric ward with the chief complaint of abdominal pain that started three days earlier.

She was seen at that time in the emergency department (ED), where she was found to have an unremarkable exam, and since the pain resolved in the ED, and her plain abdominal radiograph revealed only a large amount of stool in the colon (figure 1), she was sent home with the diagnosis of abdominal pain due to constipation. The pain continued to reoccur, and she returned to the ED on the day of her first admission. Evaluation again was thought to be abdominal pain associated with constipation, but this time the plain abdominal film revealed “scattered air-fluid levels consistent with a possible ileus or partial obstruction” (figure 2).

Her admission screening revealed no new information; specifically, there was no history of injury, nausea, vomiting, diarrhea, dysuria, flank pain or fever, and her examination revealed only normal findings including her abdomen. Her laboratory tests included a complete blood count, complete metabolic profile, amylase, lipase, urine analysis with culture, all of which were normal or negative. Because of the plain film report, she had a CT scan of her abdomen and pelvis (figures 3 and 4). The positive findings on this CT scan included a “ptotic left kidney, a bifid pelvis and a large amount of stool throughout the colon.”

She was discharged home the next day on a regimen of Miralax (polyethylene glycol). However, she was readmitted one week later with ongoing, intermittent abdominal pain. There was no new subjective information except for one low-grade fever, and her examination had not changed. Specifically, she had no spinal pain to percussion. But now, her blood work revealed a mildly elevated C-reactive protein (CRP) of 30 and erythrocyte sedimentation rate (ESR) of 59. Therefore a gastroenterology consult was obtained, and the patient underwent esophagogastroduodenoscopy and flexible sigmoidoscopy, both of which were normal. A repeat abdominal/pelvic CT scan was done, again without any significant finding, although there was mention of “hypodensity” within the fourth lumbar vertebral body (figure 5). The pediatric radiologist interpreted this as an artifact due to “volume averaging.” She again was discharged home on treatment for constipation.

You guessed it, she was readmitted nine days later for recurrent abdominal pain; only this time, the pain was aggravated by walking and range-of-motion testing of her lower extremities, and on percussion of her spine in the lower lumbar area. Her neurologic examination, specifically of the lower extremities was normal (strength, sensory and reflexes). She still had no fever, and although her CBC revealed a platelet count of 600,000, her CRP and ESR demonstrated a modest fall (14 and 48, respectively), and the rest of her blood tests were normal. An MRI of her back was performed revealing the finding in figure 6.

What’s Your Diagnosis?

1. Left renal abscess
2. Diskitis
3. Spinal epidural abscess
4. Pott’s disease

Answer

On the third hospitalization, after a three-week period of symptoms, the diagnosis of diskitis (B), involving the disk between the 4th and 5th lumbar vertebrae.

This is not an unusual way to come around to this diagnosis. These children usually present with vague pain, usually involving the pelvic area or lower abdomen, sometimes for many weeks. I’m sure you are all aware of the difficulty in localizing pain in a 2-year-old child, and this patient was no different. In her case, she did have some degree of constipation that distracted attention away from the real problem a bit, but the diagnosis will eventually demand recognition as the pain becomes more focal, involving the back at the level of the inflammation.

The L4-L5 disk is the most frequent location, but diskitis can occur at any level; it’s just much less common as one moves up the spine. In years past, the diagnosis was usually made with a bone scan, or with intervertebral disk space narrowing on plain radiographs. But nowadays, MRI is far superior with very good image detail. The MRI shown in figure 6 reveals some narrowing of the disk and abnormal enhancement adjacent vertebral bodies, probably reflecting inflammation of the vertebral end plates. The reason this condition tends to occur in young children, usually younger than 7 years of age, appears to be related to the vascular network around these structures in early childhood, which are usually obliterated as the child gets older.

Presumably, bacteremia through these vessels produces a low-grade infection, usually with Staphylococcus aureus. However, other organisms have uncommonly been recovered, such as Streptococcus pneumoniae and rarely Kingella kingae. Also, Itzhak Brook, MD, has shown that anaerobes also have a role in this entity with the first report of diskitis attributable to anaerobes (Pediatrics. 2001;7). Brook correctly points out the importance of obtaining a specimen and proper handling for recovery of aerobes and potential anaerobes. However, it seems to be getting harder to get specimens in these patients because of reluctance of our surgeons and, to a lesser extent, our interventional radiologists to go after it. It seems that patients have to show worsening on empiric therapy before a culture procedure is done. Of course the problem with that is the difficulty in recovering an organism that is “partially” treated. This is just my observation. Your experience in this regard may be quite different.

Thinking of the most common causes mentioned above, the patient was treated with IV clindamycin for a week along with bedrest, during which time she had dramatic clinical improvement, and her CRP and ESR continued to normalize. She was then discharged on oral clindamycin to complete another two to four weeks of therapy, and has remained well.

It is sometimes difficult to rule out vertebral osteomyelitis, and one must use clues in the history and physical, laboratory and imaging results available. In this case, the patient’s age is a bit young for vertebral osteomyelitis. Also, it is more common in boys. Although it may mimic diskitis, pediatric cases of vertebral osteomyelitis will usually be in older children and adolescents, presenting with chronic back pain and fever. However, it is more likely to be more acute in its presentation. MRI will also reveal vertebral body involvement rather than just end plate inflammation (figure 7). Of course, Pott’s disease is vertebral osteomyelitis caused by Mycobacteria tuberculosis. This causes chronic, progressive destruction of the vertebral body, beginning anterior and resulting in collapse of the bone, causing a gibbus deformity (figure 8, J.W. Bass collection). Of course, from a laboratory test standpoint, the inflammatory markers (CRP and ESR) would likely have continued to rise in the case of vertebral osteomyelitis.

Regarding the renal abscess choice, there was the finding of a “ptotic” left kidney, which simply means it was displaced downward, but there was no clinical or radiological evidence of a renal abscess. Sometimes when there is a “bifid pelvis,” there may be a duplicated collecting system, which may be dysfunctional and prone to infection. But that was not the case here. Plus, an abscess should be fairly easy to see on the previous imaging done, as shown in figure 9 (patient with right renal abscess).

A spinal epidural abscess is a medical and surgical emergency in most cases, with the acute onset of pain with fever and eventually cord compression symptoms. This girl had none of these findings. A typical MRI picture is seen in figure 10.

Sorry for all the radiographs, but there’s just not much to photograph with these patients, unless you want to just show the misery in their face. I could show the face of a miserable patient who could have diskitis (figure 11), but actually had something else altogether. Can you guess by looking? She will show up in a future column.

Some guest columns will be appearing soon. If you have a case with a picture, and you are interested in seeing it appear in this paper, please write to me at jhbrien@aol.com. I’ll do my best to get it in.

Commentary

A very nice, concise review of this peculiar condition known as diskitis, written by Kathleen M. Gutierrez, MD, can be found in chapter 85, pages 481–484 of the 2nd ed. of Principals and Practice of Pediatric Infectious Diseases (2003), by Sarah Long, Larry Pickering and Charles Prober. By the way, a 3rd ed. of this very informative, single-volume text will be coming out in 2007. I’ll keep you posted. Brook’s paper, mentioned above, is also referenced in this chapter.

While we are on the subject of handy references, I thought I would pass on a very nice review of empiric antibiotic use I recently found in Contemporary Pediatrics (2006;23:64–78), “What bug, which drug? Optimizing empiric antimicrobial therapy” by Echezona Ezeanolue of the University of Nevada School of Medicine in Las Vegas, Chinenye Ezeanolue, a pediatric resident at Newark Beth Israel Medical Center, Newark and Kevin Slavin of Hackensack University Medical Center and University of Medicine and Dentistry of New Jersey in Newark. It’s a common sense review along with a few clinical vignettes that I think many pediatricians and family medicine physicians would find as a useful and familiar review.