Journal Feed Weekly Wrap-Up

Spoon Feed
The prevalence of concomitant bacterial meningitis in infants 29-60 days old with UTI was 0.25%. With this low prevalence, some would argue for selective rather than routine LP in such children.

Why does this matter?
It’s unclear whether a LP needs to be performed on infants 29-60 days old found to have a UTI. What is the risk the UTI will also cause bacterial meningitis?

LP with UTI?
This was a meta-analysis of 3 prospective and 17 retrospective studies that determined the overall prevalence of bacterial meningitis in infants 29-60 days old with UTI was 0.25% (95%CI, 0.09%-0.70%). One study had 1,609 patients and made up 41.5% of the total patients in the meta-analysis. Prevalence in that study alone was 0.12% (2/1609). Of the three prospective studies, in which LP was performed on all patients, zero of a combined 72 infants had bacterial meningitis. Not all infants got a LP in the other, retrospective studies, which may have led to selection bias. However, one would think that more ill-appearing infants would be most likely to have LP performed, inflating the prevalence estimate. Regardless, the number is very low, leading to 1000 LPs to detect 2-3 cases with bacterial meningitis. Then again, it is not zero, and the finding of bacterial meningitis would certainly change duration, route (IV vs PO), and intensity of treatment. It seems a frank discussion of the actual risk with families would be in order as we discuss the risk/benefit of performing LP in these cases.

Source
Risk of Meningitis in Infants Aged 29 to 90 Days with Urinary Tract Infection: A Systematic Review and Meta-Analysis. J Pediatr. 2019 Jun 20. pii: S0022-3476(19)30536-0. doi: 10.1016/j.jpeds.2019.04.053. [Epub ahead of print]

Open in Read by QxMD

Spoon Feed
In a pregnant patient presenting with abdominal pain and suspected appendicitis, an initial ultrasound should be performed to exclude obstetric causes for abdominal pain. MRI may be a reasonable next study if it is immediately accessible and radiologists with expertise in MRI interpretation are available. Otherwise, CT should be utilized.

Why does this matter?
Pregnant patients with abdominal pain often pose a diagnostic challenge for physicians. As a result, diagnosis and treatment of appendicitis may often be delayed, resulting in increased maternal and fetal morbidity. At the same time, misdiagnosis resulting in a negative appendectomy is also associated with fetal loss.

What to do
The American College of Obstetrics and Gynecology (ACOG) and the American College of Radiology (ACR) both offer guidance on imaging in pregnancy.

• Always start with ultrasound. Establish whether there is an obstetric source of abdominal pain, as these are more common in pregnant patients.

• If MR imaging is available, get an MRI.

• If MR imaging is not available or inconclusive, get a CT. It is quick and the most definitive imaging modality.

  • Radiation exposure from CT of the abdomen and pelvis is 10-25 mGy and can be decreased to 2.5 mGy with a “pregnancy protocol CT”

  • No reported cases of fetal anomalies, growth restriction, or abortion have been reported with radiation exposure less than 50 mGy

Bottom line: Don’t withhold definitive imaging from a pregnant patient with suspected appendicitis.

Source
Diagnostic Imaging in Pregnant Patients With Suspected Appendicitis. JAMA. 2019 Jul 1. doi: 10.1001/jama.2019.9164. [Epub ahead of print]

Open in Read by QxMD

#4: CXR to Screen for Acute Aortic Syndromes

Spoon Feed
CXR was a poor screening test for acute aortic syndrome. If you suspect this diagnosis, just get a CTA.

Why does this matter?
Guidelines recommend CXR to screen for acute aortic syndromes in low pretest probability patients. As a review, the findings of acute aortic syndrome on CXR are: “poor definition or irregularity of the aortic contour, double aortic knob sign, inward displacement of aortic wall calcification by more than 10 mm, tracheal displacement to the right, displacement of a nasogastric tube, left sided pleural effusion, suspected pericardial effusion, left apical opacity,” and mediastinal enlargement (ME), “defined as a maximum width of mediastinum ≥80 mm at the level of the aortic knob or a ratio of mediastinum to chest width >0.25 or ME based on subjective evaluation.” How does CXR actually perform as a screening test?

Not the way to screen for this problem
This was a secondary analysis of ADviSED. Use of an aortic dissection detection risk score plus mediastinal enlargement on CXR had sensitivity of 67%, specificity 83%. The risk score plus any sign on CXR had sensitivity 69%, specificity 77%. For CXR alone, without the risk score, sensitivity was 54%, specificity 92% for mediastinal enlargement; sensitivity 60%, specificity 85% for any sign on CXR. Inter-rater agreement between radiologists for mediastinal enlargement was fair to moderate (k = 0.44). What this all means is that a CXR is a poor screening tool for acute aortic syndromes. If mediastinal enlargement or other signs are seen on CXR, this is concerning and needs further workup. If you suspect acute aortic syndrome, it’s best to just get a CTA.

Source
Integrated use of conventional chest radiography cannot rule out acute aortic syndromes in emergency department patients at low clinical probability. Acad Emerg Med. 2019 Jun 20. doi: 10.1111/acem.13819. [Epub ahead of print]

Open in Read by QxMD