Methods
A systematic search of PubMed, EMBASE and Cochrane library was done with
no restriction on publication date. The following search strategy was
used: [infectious mononucleosis OR glandular fever] AND [liver
function* OR liver enzyme* OR ultrasound]. Two authors (ETT, DW)
independently screened title and abstract according to prespecified
criteria and full texts of relevant articles were retrieved for further
eligibility assessment. References of included full text articles were
further screened for relevant studies. Any discrepancies on study
eligibility were discussed with a third author (OE) and resolved
accordingly.
We included all original studies evaluating any form of LFTs and/or
abdominal ultrasound (USS) in both adult and/or paediatric patients with
suspected or confirmed diagnosis of IM. Standard LFTs included any of
the following: Aspartate Transaminase (AST), Alanine Transaminase (ALT),
Alkaline Phosphatase (ALP), Bilirubin, Albumin, Total protein and
Gamma-Glutamyltransferase (GGT). Abdominal ultrasound included liver
and/or spleen and/or gallbladder ultrasound. We excluded studies
evaluating patients with solely viral hepatitis (i.e. not related to
clinical presentation of infectious mononucleosis), reviews, non-human
studies, conference abstracts, letters/commentaries, expert opinions or
recommendations, case reports and studies not published in English.
A standardized data collection proforma was used. This was initially
piloted on five studies by all authors and revised accordingly.
Subsequent data collection was by one author (DW) and checked for
accuracy by a second author (ETT). In case of duplicate publications, we
included the manuscript with the most complete data. The following data
items were extracted: study characteristics, participants’ demographics,
method of ascertainment of IM, evaluation of LFTs, clinical findings on
abdominal ultrasound and complications relating to liver disease. Our
main outcomes are the proportion of patients with abnormal LFTs at
presentation and time to resolution of LFTs. Secondary outcomes include
presence of clinical hepatosplenomegaly, findings of abdominal
ultrasound and any evidence of decompensated liver disease.
Quality assessment of included studies were performed using Murad’s
tool.3Although the tool was originally developed for
case series and case reports, this was discussed among the authors and
deemed suitable for our review irrespective of study type as we were
only extracting data relating to a single arm of interest. Data was
expressed as count and percentages. Proportion of abnormal LFTs in
individual studies were aggregated and an average calculated for all
studies.
PRISMA reporting guideline was used in preparation of the manuscript.