I hope that you have had a go at trying to work out this case. This month, Prof Matsumura (Japan) and Prof Dhaliwal (USA) have kindly contributed their expert opinions to this case. A contribution was also gratefully received from a first year Japanese resident (shown below). As a reminder, here are the original questions to the case:
Question 1: From the history, physical exam, limited laboratory and radiology data, what is your differential diagnosis?
Question 2: What emergency radiology test will you do next?
Question 3: Where is the potential origin of this problem?
Question 4: What emergency treatment(s) will you consider in this patient?
From Professor Matsumura: Thank you very much for showing me a challenging case again.
This 65-year-old lady had fever and back pain. This combination is red flag sign. Back pain is very common condition (I have had this before!). However, she had fever too. I should think infection or neoplastic diseases. Moreover, she had had history of poorly controlled diabetes mellitus and bilateral renal calculi. I think these are very important information.
Question 1: From the history, physical exam, limited laboratory and radiology data, what is your differential diagnosis?
Problem list is as follows:
#1 Back pain
#4 Poorly controlled diabetes mellitus
#5 Renal calculi
#6 History of UTI
#8 Low blood pressure
#9 Globular distension and tender in suprapubic area
#10 Unable to grip the examiners finger with her anus.
#11 Weakness in the lower limbs
#12 Absence of knee jerks
#13 Tender spine and muscles from L1-L5 region
#14 Psoas sign
#16 Septicemia, E. coli and Klebsiella pneumoniae
#17 Gas in the soft tissue in an abdominal CT scan Differential diagnoses are as follows.
Vascular: Less likely
Infection: Emphysematous pyelonephritis, Discitis, Osteomyelitis, Epidural abscess
Neoplastic: Lymphoma, Metastatic tumor
Autoimmune: Less likely
Trauma/Degenerative: Less likely
Iatrogenic: Less likely Idiopathic: Less likely
Congenital: Less likely
#1 Back pain, #2 Fever, #3 Dysuria, #4 DM, #5 Renal calculi, #6 History of UTI, #9 Globular distension and tender in suprapubic area.
Urinary tract infection especially pyelonephritis is suspected.
#11Weakness in the lower limbs, #13 Tender spine and muscles from L1-L5 region, #14 Psoas sign. These conditions indicate inflammation spreads to retroperitoneal area including psoas muscle.
#17 Gas in the soft tissue in an abdominal CT scan. I think emphysematous pyelonephritis is most likely.
#7 Tachycardia, #8 Low blood pressure, #15 leukocytosis, #16 Septicemia, E. coli and Klebsiella pneumoniae
These findings indicated sepsis. E. coli and Klebsiella pneumoniae are major organisms in emphysematous pyelonephritis.
#10 Unable to grip the examiners finger with her anus Rule out cauda equina syndrome. #12 Absence of knee jerks. This patient had diabetes. She can have diabetic neuropathy.
Question 2: What emergency radiology test will you do next? I would perform MRI of lumbar spine to exclude cauda equina syndrome, osteomyelitis, or discitis.
Question 3: Where is the potential origin of this problem? I think urinary tract infection is likely.
Question 4: What emergency treatment(s) will you consider in this patient? If we can see typical images of emphysematous pyelonephritis, surgical resection of the involved tissue in addition to systemic antimicrobial therapy is needed.
From Dr Dhaliwal: This is a very interesting case. I have deviated from the four question format instead by “thinking aloud” as the case unfolds. My differential diagnoses are embedded throughout with diagnostic and treatment suggestions at the conclusion.
History: Fever and low back pain typically brings to mind an infectious process. The low back pain may reflect nonspecific myalgias (e.g., influenza, SBE) or can be a focus of a suppurative retroperitoneal or lumbar spine infection. The difficulty walking, standing, and urinating points to a neuraxial infection, such as epidural abscess, vertebral osteomyelitis, or discitis. Poorly controlled DM increases the susceptibility to infection in general and raises the possibility of specific complicated infections, e.g., emphysematous cystitis or pyelonephritis. The latter must be entertained with pre-existing renal calculi. She lacks the dysuria that typically accompany lower tract infection.
Exam: This is an ill patient in shock. A murmur to suggest endocarditis is not detected, but can be challenging to hear at a heart rate of 120. The tender distended suprapubic area is compatible with complicated cystitis. A pulsatile mass – which would raise suspicion of one of the most sinister fever and low back pain entities, mycotic aneurysm – is absent. The anal laxity and decreased lower extremity tone and strength collectively suggest a spinal cord syndrome, perhaps of the cauda equina variety. The areflexia is consistent with nerve root compression or early direct cord compression. Interestingly, there does not appear to be a sensory level that accompanies cord syndromes, although the sensory exam can be highly variable and subjective at times. The musculoskeletal exam appears to suggest a psoas sign of sorts, which raises the possibility of a psoas abscess, along with inflammation of any structure adjacent to the psoas.
Labs/Radiology: Gram negative bacteremia suggests GI or GU pathology. Polymicrobial bacteremia should at least trigger the thought that a connection that has been established between the GI tract and the vasculature such as an aortoenteric fistula, an ingested foreign body (e.g., toothpick), or an eroding cancer. The CT finding of soft tissue gas suggests a gas-forming organisms (like E coli and Klebsiella) and is compatible with emphysematous cystitis or pyelonephritis. Mesenteric infarction or Fourneir’s gangrene could account for soft tissue gas as well, but supporting clinical features are lacking.
Conclusion: 65 year old teacher with poorly controlled DM with E Coli and Klebsiella sepsis, low abdominal/back pain, and lower extremity weakness. One pressing question is whether there is a space occupying lesion compressing her spinal cord/roots. Numerous elements of the case suggest this, and therefore I would be most interested in a lumbosarcral spinal MRI. The next question (regardless of the MRI results) would be the origin of her gram negative bacteremia. Examination of the urine (with catheter placement) and re-examination of the CT scan would be in order. Treatment pending the above studies would include vigorous fluids and IV antibiotics.
- Emphysematous cystitis
- GNR bacteremia
- Epidural abscess
Among DIET IN HIM, Infection is the most likely cause of the illness, b/c DM, B/C results, acute course of the symptoms etc. strongly suggest Infection. Also I would consider Neoplasm b/c it could be similar to infection even though it is less likely, given back pain day-to-day progressing and neurological exams. I don't have ideas of neurological and inflammation cause in this case b/c neurological exams show bilateral but motor-only impairments and her age.
Secondly, I think this patiet have DM neuropathy symptoms such as rectovesicle impairment but it does not explain bilateral pure motor weakness, maybe. In this case, I am confused by this bilateral proximal limb weakness, but it could be explained by infection such as abscess b/o the presence of poas signs. Anyway, DDx as follows.
#1 epidural abscess #2 discitis #3
lymphoma and other cancers #4 DM neuropathy #5 infection of the soft tissue it might cause the bacteremia.
Q2 Enhanced MRI of the lumbar spine if available soon. If not, enhanced CT is also a good choice of study.
Q3 Vertebra~Disc of L3-5
Q4 ABx covering at least GNR (especially E.coli and Klebsiella). But I consider broad ABx use until Dx is determined.
Both Professors Matsumura and Dhaliwal's general diagnostic process was similar to my own so to save time, I shall not repeat another list of my own. However, the history of renal calculi and diabetes mellitus are extremely important in this case. They are risk factors for infection. As has already been alluded to, fever and back pain together are a Red Flag sign and such patients need urgent investigation. Serious diagnoses such as osteomyelitis, discitis, epidural abscess and psoas abscess should be high on the differential diagnosis list.
Moreover, with the addition of neurological signs such as an inability to pass urine, weakness of proximal musculature, and inability to grip the examiners finger during rectal examination all point to a possible cord / cauda equina syndrome as both Professors have already mentioned.
Indeed, an emergency MRI of the lumbosacral region would be warranted.
When this patient was seen by another senior doctor several days after admission, the weakness of the lower limbs had progressed more distally. The patient could only move her toes and there was no movement against gravity. Interestingly, the patient still denied sensory changes in her lower limbs.
The CT scan revealed an obstructed right kidney and a normal left kidney. There was a mild 'dirty fat' sign around the right kidney and evidence of bilateral psoas abscesses with gas formation throughout the muscle plains. This gas extended deep into the erector spinae musculature. The liver, gallbladder, pancreas, and bowel all appeared normal (no toothpicks).
The infection appeared extensive throughout the lumbar area. Emergency MRI was performed and there was no evidence spinal cord or cauda equina compression. Unfortunately, no urine analysis data was available, which in retrospect, could have been extremely useful.
The suspected diagnosis was of an emphysematous pyelonephritis.
The patient was given ceftriaxone and underwent emergency surgery which included:
- Debridement of the erector spinae muscles; the wound was left open and examined and washed daily.
- Insertion of drains into both psoas muscles to drain the abscesses, again with daily wash outs.
After 6 weeks of intravenous antibiotics and surgical lavage therapy, the patient had improvement in well-being, in general blood parameters and became afebrile.
- Final Diagnosis: Emphysematous Pyelonephritis (Class 3b)
This is an extremely serious infection. In previously described cases, most patients have been women aged over 60 years with a history of diabetes and / or urinary tract obstruction. Usual organisms include E. coli and K. pneumoniae. Candida is also a rare cause.
Causes of urinary tract obstruction can be either papillary necrosis or less commonly, renal calculi.
The symptoms of emphysematous pyelonephritis cannot be separated from those of the more usual pyelonephritis. Patients complain of nausea, vomiting, fevers, rigors, and flank or abdominal discomfort. The onset of symptoms may occur suddenly or evolve slowly over a several weeks.
Gas can sometimes be elucidated on the plain abdominal film but CT is more sensitive and specific.
The infection is managed with intravenous antibiotics. Some cases can be successfully treated with antibiotics and percutaneous drainage of pus from the kidney with relief of the obstruction. Failing that, open surgery is required with nephrectomy.
Emphysematous pyelonephritis can be classified as follows:
- Class 1: Gas in the collecting system only (ie, emphysematous pyelitis)
- Class 2: Gas in the renal parenchyma without extension to the extrarenal space
- Class 3A: Extension of gas or abscess to the perinephric space, which is defined as the area between the fibrous renal capsule and the renal fascia
- Class 3B: Extension of gas or abscess to the pararenal space, which is defined as the space beyond the renal fascia and/or extension to adjacent tissues such as the psoas muscle
- Class 4: Bilateral emphysematous pyelonephritis or a solitary functioning kidney with emphysematous pyelonephritis.
Nevertheless, despite intensive treatment, the mortality from several studies can be between 7 to 19%. Not numbers to be ignored.
Neurological impairment The physical examination of hypotonia, weakness and loss of reflexes is consistent with a neurological component. Despite the apparently normal MRI scan on admission, with worsening neurology post-operatively, it would be judicious to repeat the MRI scan and perform nerve conduction studies.
Although, direct muscle damage due to infection, and as part of surgical intervention, can result in weakness, particularly of the postural muscles of the spine and flexors of the thigh (psoas) in this patient, it does not account for the distal weakness or loss of reflexes alone.
It is possible that the infection resulted in nerve damage, possibly due to diffuse radiculoneuropathy. However, a cauda equina syndrome would still need to be re-excluded by repeating an MRI. The fact that the patient stated that there was no sensory deficit is intringuing. I would agree with Prof Dhaliwal that the physical examination can be subjective on occasion although I would like to add that such an exam should be done in a standardised and methodical way to ensure that all sensory modalities are checked. Diabetic neuropathy can easily be missed unless it is specifically looked for, and the patients sometimes do not even realise that they have a loss in sensation. I would invite others to comment on this point.
The fact that she had existing diabetes can indeed account for loss of reflexes which should not be forgotten. Only with further detailed repeat neurology (in full) and repeat MRI plus nerve conduction studies can the cause and full the extent of the disease be appreciated. The cause portends the prognosis.
Learning Points from This Case
- Don't ignore Back Pain and Fever -- together they are a Red Flag Sign -- a potential medical emergency
- Make an assessment and consider the serious diagnoses that should not be missed e.g. abscess, discitis, osteomyelitis etc.
- If a patient complains of leg weakness then take it seriously especially when there is a history of back pain and fever. It suggests a possible cord / cauda equina compression - a neurosurgical emergency. Do not send the patient home : urgent physical examination, lab studies and radiology (CT/MRI) are required; this is especially important with neurological impairment
- Gas in the tissues is serious. Antibiotics, drainage of the kidney, percutaneously and urethral catheterisation to relieve obstruction are advised. Surgery may also be required depending on the extent of the disease - see the classification above.