Low hemoglobin-albumin-lymphocyte-platelet score as a risk marker for stroke-associated pneumonia in acute ischemic stroke: a retrospective cohort study.
Wu Z, Sun Q, et al. • Frontiers in neurology • 2026
Low HALP score was independently and non-linearly associated with stroke-associated pneumonia in acute ischemic stroke, with an exploratory threshold near 27.7 below which each log2-unit increase in HALP was associated with a 53% lower SAP risk.
Key Findings
Results
Stroke-associated pneumonia occurred in 21.3% of acute ischemic stroke patients in this cohort.
1,766 consecutive AIS patients were admitted between January 2019 and December 2023.
SAP occurred in 376 patients (21.3%).
HALP was calculated from blood tests obtained within 48 hours of admission.
This was a retrospective cohort study design.
Results
HALP scores were significantly lower in AIS patients who developed SAP compared to those who did not.
Median HALP in SAP patients: 35.08 (IQR 18.59–51.86).
Median HALP in non-SAP patients: 44.55 (IQR 31.29–63.25).
The difference was statistically significant (p < 0.001).
HALP score incorporates hemoglobin, albumin, lymphocyte, and platelet values and reflects nutritional, inflammatory, and immune status.
Results
Each 1-unit increase in log2-HALP was independently associated with a 29% lower risk of SAP after full multivariable adjustment.
OR = 0.71, 95% CI: 0.62–0.83, p < 0.001.
Association was assessed using multivariable logistic regression.
This result was obtained after full adjustment for potential confounders.
Results
Patients in the highest HALP quartile had significantly lower SAP risk compared to those in the lowest quartile.
OR = 0.62, 95% CI: 0.41–0.95 for highest vs. lowest quartile.
P for trend = 0.018 across quartiles.
This indicates a graded inverse relationship between HALP and SAP risk.
Results
A non-linear association was identified between HALP and SAP risk, with an exploratory threshold at approximately HALP = 27.7.
Non-linearity was identified using generalized additive modeling and two-piecewise logistic regression.
Below the threshold of ~27.7, each log2-unit increase in HALP was associated with a 53% lower SAP risk (OR = 0.47, 95% CI: 0.35–0.63, p < 0.001).
Above the threshold of ~27.7, no significant association between HALP and SAP was observed.
The threshold value of 27.7 is described as 'exploratory.'
Results
Adding HALP to the A2DS2 score produced a statistically significant but modest improvement in SAP discrimination.
AUC for A2DS2 alone: 0.758.
AUC for A2DS2 combined with HALP: 0.779.
The improvement was statistically significant (p = 0.004).
The authors characterized the improvement as 'statistically significant but modest.'
What This Means
This research suggests that a blood-test-based score called the HALP score — which combines measures of hemoglobin (related to red blood cells), albumin (a protein reflecting nutrition), lymphocytes (immune cells), and platelets (clotting cells) — is linked to the risk of developing pneumonia after a stroke. Among 1,766 stroke patients studied over five years, about 1 in 5 developed stroke-associated pneumonia, and those who did had notably lower HALP scores than those who did not. Even after accounting for other health factors, patients with lower HALP scores were significantly more likely to develop pneumonia following their stroke.
An important nuance the researchers found is that this relationship is not simply a straight line — it is non-linear. The protective association of a higher HALP score appeared most strongly in patients whose HALP was below approximately 27.7; above that level, higher HALP scores did not appear to offer additional protection against pneumonia. This suggests that patients with very low HALP scores at admission may represent a particularly high-risk group worth monitoring closely. Adding HALP to an existing stroke-pneumonia risk tool (the A2DS2 score) modestly improved its ability to identify which patients would go on to develop pneumonia.
This research suggests that routine blood tests taken shortly after stroke admission could be used to calculate a HALP score and potentially help clinicians identify patients at higher risk for this serious complication. Because this was a retrospective study at a single center, the findings — particularly the specific threshold value of 27.7 — need to be validated in future, independent studies before they could inform clinical practice.
Wu Z, Sun Q, Zhan J, Yang M, Duan T, Pan H, et al.. (2026). Low hemoglobin-albumin-lymphocyte-platelet score as a risk marker for stroke-associated pneumonia in acute ischemic stroke: a retrospective cohort study.. Frontiers in neurology. https://doi.org/10.3389/fneur.2026.1866716