Asthma affects roughly 363 million people worldwide, including an estimated 45.7 million adults in China. Standard treatment built on inhaled corticosteroids controls the disease for most patients, yet a stubborn minority remain poorly controlled despite it — and long-term, high-dose inhaled steroids carry their own costs, from hoarseness and oral thrush to concerns about bone and adrenal effects.

A systematic review and meta-analysis published in Frontiers in Medicine on November 19, 2026, takes one such adjunct — Huanglong Zhike Granules, a marketed Chinese patent medicine — and asks the question properly. It pools 17 randomised controlled trials involving 1,583 patients, grades the certainty of the evidence, and is unusually frank about what the pooled numbers cannot support. It is worth reading precisely because its headline result looks encouraging and its fine print says: not so fast.

What the Formula Is

Huanglong Zhike Granules (HZG) is a modern patent medicine manufactured in China, composed of nine medicinal materials: Astragali Radix (Huangqi), Pheretima (earthworm), Epimedii Folium, Platycodonis Radix (Jiegeng), Belamcandae Rhizoma, Houttuyniae Herba, prepared Ephedrae Herba, Crataegi Fructus, and Descurainiae Semen. In traditional terms it is used to replenish qi and tonify the kidney, clear the lung, and relieve cough — and it is used mainly for asthma, particularly cough variant asthma, the form in which a persistent cough is the dominant symptom.

Chemical analysis of the granules has identified about 20 constituents, including astragaloside IV from Huangqi, icariin from Epimedii Folium, platycodin D from Jiegeng, and ephedrine from prepared Ephedra. Animal studies suggest the formula can reduce airway hyperresponsiveness and lung inflammation in experimental asthma. That is biological plausibility, not proof — and the review's authors say so themselves. The clinical question is separate, and it is the one the meta-analysis addresses.

What the Review Did

The review was registered in advance with PROSPERO and reported to PRISMA 2020 standards. The authors searched seven databases — PubMed, Web of Science, the Cochrane Library, Embase, and three Chinese databases — from inception to October 27, 2026, and included only randomised controlled trials in which patients received the granules in addition to conventional Western therapy, against the same conventional therapy alone. Two reviewers extracted data independently, assessed risk of bias with the Cochrane RoB 2 tool, and graded the certainty of evidence with GRADE.

That design matters. Because the granules were always tested as an add-on, the review can only speak to adjunctive use. It says nothing about the formula as a stand-alone treatment, and nothing about stopping prescribed inhalers — a distinction worth keeping in mind, because it is exactly the distinction that gets lost when a positive result travels from a journal to a headline.

What It Found

Across the five primary outcomes, the pooled results favoured the combination therapy:

  • FEV1 (the volume of air exhaled in one second) improved by 0.29 litres on average, across 10 trials.
  • FVC (forced vital capacity) improved by the same 0.29 litres, across 8 trials.
  • FEV1/FVC rose by 6.07 percentage points, across 9 trials.
  • Peak expiratory flow improved with a standardised mean difference of 1.13, across 10 trials — a large standardised effect, though standardised effects are harder to translate into everyday terms than litres are.
  • Overall response rate — the proportion of patients classed as responders by each trial's own criteria — was 20% higher in relative terms (risk ratio 1.20), across 16 trials, and this was the one primary outcome with no statistical heterogeneity between studies.

Secondary outcomes pointed the same way. Markers of airway inflammation (TNF-α, IL-6, and exhaled nitric oxide) fell, as did markers of type-2 inflammation (IL-5 and eosinophils) and an indicator of airway remodelling (MMP-9). Airway wall thickness and wall area were also lower in the combination group, though only two or three trials contributed those measures.

But one result did not follow the pattern. The Asthma Control Test, the patient-reported measure of how well asthma is actually controlled day to day, showed no statistically significant difference between groups. Only two trials reported it — a small, telling crack in an otherwise uniformly positive picture: lung-function numbers moved, but the measure closest to how patients feel did not clearly move with them.

How to Read the Numbers

A pooled effect is only as trustworthy as the trials behind it, and this is where the review earns its keep. Four problems recur throughout the evidence base.

First, the certainty of evidence was low for all five primary outcomes. GRADE low certainty means the true effect may be substantially different from the pooled estimate. The downgrades came from risk of bias, inconsistency between studies, and imprecision — the three most common reasons a meta-analysis of small trials gets a low grade.

Second, the trials themselves were methodologically weak. All 17 raised concerns about missing outcome data. Sixteen of 17 raised concerns about deviations from the intended interventions and about selective reporting. Four were judged at high risk of bias in how outcomes were measured. No trial in the set was clean across all five RoB 2 domains.

Third, heterogeneity was substantial and largely unexplained. I² values of 78–88% for the lung-function outcomes mean the trials disagreed with one another far more than chance would predict — different populations, background treatments, durations, and in some cases modified versions of the formula. Subgroup analyses hinted that typical bronchial asthma responded more than cough variant asthma, but only two trials fed the typical-asthma subgroup, so that finding is fragile.

Fourth, the evidence base is narrow. All 17 trials were conducted in China, most enrolled children or patients with cough variant asthma, and treatment lasted at most 12 weeks — so the results say least about adults with typical bronchial asthma, and nothing about long-term use. For the overall response rate, Egger's test also detected possible small-study effects, a hint that less favourable trials may be missing from the published record. A trim-and-fill adjustment still left the effect significant, but that is an exploratory correction, not a guarantee.

The Safety Question

Ten of the 17 trials reported adverse events: 23 events in the combination group and 26 in the control group, a pooled risk ratio of 0.89 with a wide confidence interval (0.52–1.51) that is compatible with both a reduction and an increase in risk. Reported reactions included dry mouth, hoarseness, drowsiness, nausea, headache, palpitations, diarrhoea, and dizziness. No serious safety signal emerged — but with at most 12 weeks of follow-up, incomplete reporting, and fewer than 50 events in total, the review cannot establish that the granules are as safe as standard therapy alone, let alone safer.

One specific caution deserves attention. The formula contains prepared Ephedra, and ephedrine is among its identified constituents. Ephedrine and related alkaloids can raise heart rate and blood pressure, and their effects could add to those of inhaled β₂-agonists — the reliever inhalers many asthma patients use. The included trials did not collect the data needed to quantify that risk. It is a reason for monitoring, not for alarm, but it is also a reason the safety chapter of this story is unfinished.

The Practical Reading

For a reader with asthma, the honest summary is this: in 17 small, mostly short trials, all conducted in China, adding Huanglong Zhike Granules to standard therapy was associated with better lung-function numbers and lower inflammatory markers, with no clear difference in reported side effects over 12 weeks — and with no clear improvement in the patient-reported measure of asthma control. The evidence is graded low certainty, which means the true effect could be smaller, or larger, than the pooled numbers suggest.

That is not a verdict against the formula. It is a verdict about the state of the evidence, and it points directly at what the next study should look like: larger, multicentre, preregistered, placebo-controlled, with standardised dosing and background therapy, with exacerbation rates, rescue-medication use, and quality of life as outcomes — and with systematic monitoring of blood pressure, heart rate, and sleep, especially in children. Until such a trial exists, the granules belong in the category of "promising adjunct, unproven" — a category traditional medicine fills often, and one that deserves better evidence rather than louder claims.

There is a broader lesson here too, and it is one this site returns to often. A meta-analysis is not a stamp of approval; it is a way of asking, of 17 trials at once, how much they can collectively support. When the answer is "something, but not much, and not for long," that is still useful knowledge — it tells the patient what to ask their clinician, and it tells the researcher exactly which trial to run next. The worst use of a positive meta-analysis is to treat low-certainty evidence as a settled fact. The best use is to treat it as a well-drawn map of what we do not yet know.


Source: Sun Y, et al. "Efficacy and safety of Huanglong Zhike Granules as adjunctive therapy for asthma: a systematic review and meta-analysis." Frontiers in Medicine. Published November 19, 2026. DOI: 10.3389/fmed.2026.2005189. PROSPERO registration: CRD420261488793.

If you are interested in how the classical Chinese medical tradition thinks about health more broadly — its ideas of balance, prevention, and the body's patterns — our book Huangdi Neijing: A Modern Introduction offers a beginner-friendly guide to the foundational text of TCM. For another example of how to read a large body of herbal trials without being misled by it, see our evidence review of 33 trials of Chinese herbs for dry eye.

This article is for educational purposes only and is not medical advice. If you have asthma, do not change your treatment without consulting a qualified healthcare professional.