Blood-Based Inflammation and Nutrition Scores Reveal Immune Imbalance in Psychiatric Patients Testing Positive for Drugs

Three simple numbers calculated from a routine blood draw may offer a new window into the biology of psychiatric illness and substance use. A retrospective study published in BMC Psychiatry reports that psychiatric inpatients whose urine tested positive for drug metabolites showed a higher systemic inflammation response index and a lower prognostic nutritional index than healthy controls, while a third measure, the systemic immune-inflammation index, failed to separate the groups. The findings, drawn from more than a thousand participants treated at a single Turkish city hospital between 2022 and 2024, suggest that low-cost hematological markers could eventually help clinicians gauge immune and nutritional dysregulation in psychiatric care, although the authors caution that the signals are not uniform across diagnoses or substances.

The research team, led by Ayla Yildiz of the Department of Medical Biochemistry at Başakşehir Çam and Sakura City Hospital in Istanbul, together with Binnur Karabiyik from the hospital’s psychiatry department and Levent Deniz from the University of Health Sciences Türkiye Istanbul Training and Research Hospital, set out to answer a question that had not previously been addressed in the literature: whether indices derived from standard blood counts differ between psychiatric patients with documented drug exposure and people without psychiatric illness. Because these indices require nothing more than a complete blood count and, in one case, routine biochemistry, they have attracted growing interest across medicine as accessible proxies for systemic inflammation and nutritional status.

The three indices at the heart of the study each capture a different facet of immune balance. The systemic inflammation response index, or SIRI, is computed from the counts of monocytes, neutrophils, and lymphocytes, and is interpreted as a reflection of the innate inflammatory response. The systemic immune-inflammation index, or SII, multiplies the platelet count by the neutrophil-to-lymphocyte ratio, combining clues about clotting activity and cellular immune balance into a single figure. The prognostic nutritional index, or PNI, blends the absolute lymphocyte count with serum albumin concentration, and is conventionally used as a measure of immunonutritional reserve. Elevated values on the first two indices are generally read as signs of heightened systemic inflammation, whereas lower PNI values point toward depleted nutritional and immune reserves.

To test whether these markers behave differently in the presence of psychiatric illness and drug use, the investigators assembled three groups. The largest comprised 760 psychiatric inpatients whose urinary drug metabolite screening was positive. A second group consisted of 171 psychiatric patients without positive metabolites, and the third included 151 healthy controls. Patients with infectious, autoimmune, or malignant diseases, or with incomplete data, were excluded, a step designed to prevent conditions known to distort inflammatory and nutritional markers from confounding the comparison. All indices were calculated from hematological and, where relevant, biochemical parameters collected as part of routine care.

The headline result was a clear divergence on two of the three measures. Patients carried a median SIRI of 1.24, with an interquartile range of 0.79 to 1.97, compared with 0.98, ranging from 0.74 to 1.27, in controls. Their mean PNI was 56.0, with a standard deviation of 6.12, against 58.8, with a standard deviation of 3.87, in the healthy group. Both differences were statistically significant at p less than 0.001. The SII, by contrast, did not differ significantly between patients and controls, with a p value of 0.108, indicating that the combined platelet and neutrophil-lymphocyte signal was not the discriminating variable in this population.

Regression analyses reinforced the pattern. After adjusting for age and sex, SIRI remained positively associated with the presence of psychiatric disease, with an odds ratio of 1.847 and a 95 percent confidence interval spanning 1.423 to 2.398, meaning each unit increase in the index was tied to nearly a doubling of the odds of belonging to the patient group. PNI moved in the opposite direction, showing an inverse association with disease presence, with an odds ratio of 0.917 and a 95 percent confidence interval of 0.889 to 0.947. In other words, higher inflammation readings and lower nutritional reserve scores both tracked with psychiatric illness in this cohort.

Receiver operating characteristic analysis, which evaluates how well a marker separates two groups across all possible cutoffs, added a diagnostic dimension. In the subgroup of drug-positive patients with schizophrenia, the area under the curve reached 0.727 for SIRI and 0.804 for PNI. An area under the curve of 0.5 indicates performance no better than chance, while 1.0 indicates perfect separation, so values in the 0.7 to 0.8 range are generally considered moderately discriminative. The PNI figure in particular suggests that the combined lymphocyte-albumin score carried meaningful information about which individuals belonged to the drug-positive schizophrenia subgroup in this sample.

Yet the study’s most instructive findings may be the ones that did not survive scrutiny. Once the researchers adjusted for age and sex and applied corrections for multiple comparisons, no significant drug-status-related differences in SIRI or SII remained within the diagnostic subgroups. The one signal that persisted was a lower PNI among drug-positive patients with nonorganic psychosis. This pattern of attenuation matters because it highlights how easily raw group differences in inflammatory markers can reflect demographic composition rather than biology, and it underscores the need for careful statistical control in biomarker studies of psychiatric populations.

The authors also observed that substance-specific associations varied according to both psychiatric diagnosis and the type of substance involved, meaning that the relationship between drug exposure and immune-nutritional markers was not a one-size-fits-all phenomenon. Urinary drug metabolite screening in the study covered substances such as tetrahydrocannabinol, the principal psychoactive component of cannabis, and amphetamine-class compounds, and the interplay between a given substance and a given diagnosis appeared to shape the biomarker profile. This heterogeneity cautions against interpreting any single index as a universal signature of drug-related immune dysregulation in psychiatry.

Taken together, the study establishes that differences in SIRI and PNI exist between psychiatric patients and healthy controls, but it equally demonstrates that these differences are inconsistent across diagnostic and substance-use subgroups and that some associations weaken substantially after adjustment for age and sex. The researchers conclude that the variability in inflammatory and immunonutritional indices across psychiatric and substance-use subgroups warrants further investigation. For now, the work positions routine blood-derived indices as promising, inexpensive candidates for monitoring immune and nutritional status in psychiatric care, while making clear that their clinical value will depend on understanding how diagnosis, substance type, and demographic factors jointly shape what these numbers mean.

Subject of Research: Inflammatory and nutritional blood indices in psychiatric patients with positive urinary drug metabolites

Article Title: SIRI, SII, and PNI in psychiatric patients with positive urinary drug metabolites: association with immune and nutritional biomarkers

Article References: SIRI, SII, and PNI in psychiatric patients with positive urinary drug metabolites: association with immune and nutritional biomarkers. (n.d.). https://doi.org/10.1186/s12888-026-08587-9

Image Credits: AI Generated

DOI: 10.1186/s12888-026-08587-9

Keywords: SIRI, SII, PNI, psychiatry, drug metabolites, inflammation, biomarkers, schizophrenia, substance use, nutritional status, hematology, retrospective study