The most direct evidence is preclinical. Choi and colleagues (2012)
applied a cream containing guava-leaf water extract to
dinitrochlorobenzene-induced eczema-like lesions in NC/Nga mice and
reported lower lesion scores, IgE, TARC and epidermal thickening [4]. Han
and colleagues (2011) showed the extract suppressed the Th2 chemokine
TARC in human keratinocytes via heme oxygenase-1 induction and NF-κB/STAT1
blockade [5] — human cells, but a dish, not a trial. Ojewole (2006)
reported reduced experimentally induced paw swelling and pain responses in
rodents given the aqueous leaf extract [6], and Qa'dan and colleagues
(2005) found the leaf extract inhibited organisms cultured from acne
lesions — an in vitro study on clinical isolates, not a treatment trial
[7].
A consistent preclinical picture, a documented traditional practice, and
no human trial of the topical use: grade C, with the mouse-to-human
translation as the central open question.
Human trials of guava leaf do exist, though none of them tests the use
this monograph describes. The landmark study is a clinical trial of a
standardised leaf phytodrug in acute diarrhoeic disease [10] — the leaf's
headline traditional indication, and the anchor everything since cites. A
randomised, double-blind pilot reported effects of a leaf extract in
Japanese subjects with knee pain [11], and a placebo-controlled trial
tested a guava-leaf mouthrinse as part of an oral-care regimen [12]. A
topical formulation has been studied for thermoregulatory effect after
exercise [13]. A 2024 review synthesises the antidiarrhoeal literature
across forty-four countries and maps it onto the identified bioactives
[14].
So the ladder reaches the clinical rung for the leaf — but for the gut and
the mouth, not for irritated skin. The grade stays at C because the
question this entry asks remains untested in humans.
