Multiple myeloma is a cancer of plasma cells, the antibody-producing white blood cells that reside in the bone marrow. Although novel therapies have transformed outcomes for many patients over the past two decades, those who deteriorate to the point of requiring intensive care face a grim prognosis, and clinicians have long lacked reliable tools to determine, within the first hours of an ICU stay, which patients warrant the most aggressive interventions. The new study addresses this gap by examining whether blood urea nitrogen, commonly abbreviated BUN, carries prognostic weight in this specific setting. BUN has already been established as an independent prognostic factor in a range of other diseases, but its significance in critically ill patients with myeloma had remained unclear until now.

To investigate the question, the researchers turned to the Medical Information Mart for Intensive Care, version 2.2, a widely used database known as MIMIC-IV that contains de-identified health records of thousands of patients treated at Beth Israel Deaconess Medical Center in Boston. The database, jointly maintained by the Massachusetts Institute of Technology and the medical center, allowed the team to identify adult patients whose first intensive care unit admission carried a diagnosis of multiple myeloma. The authors measured BUN levels within the first 24 hours of that admission and then tracked patient outcomes over three time horizons: 28-day mortality served as the primary endpoint, while 90-day and 365-day mortality were analyzed as secondary outcomes. Because the database was established under ethics approvals from MIT and Beth Israel Deaconess Medical Center with consent obtained during the original data collection, the study was exempt from additional ethical approval and informed consent requirements.

The statistical machinery behind the analysis was rigorous. The team employed Cox proportional hazard regression models, the standard epidemiological tool for estimating how a continuous variable such as BUN modifies the hazard of death over time while adjusting for confounding factors, alongside Kaplan-Meier survival analyses, which graphically depict how survival probabilities diverge between patient groups. Subgroup and sensitivity analyses were conducted to probe whether the association held consistently across different patient characteristics and modeling choices, a critical step in retrospective database research where hidden biases can distort apparent relationships.

The headline result is striking in its simplicity. After full statistical adjustment, every 5 milligrams per deciliter increase in BUN was associated with a 6 percent increase in 28-day mortality, corresponding to a hazard ratio of 1.06 with a 95 percent confidence interval of 1.01 to 1.11. The confidence interval, which excludes the null value of 1.0, indicates that the association is statistically significant and unlikely to be a product of random variation alone. When the researchers categorized patients by BUN level, the contrast became even more dramatic: patients whose BUN exceeded 43 milligrams per deciliter faced a hazard ratio of 6.36 for 28-day mortality, with a confidence interval of 2.27 to 17.78, compared with the reference group whose BUN levels were below 24 milligrams per deciliter. In practical terms, a critically ill myeloma patient with markedly elevated BUN at admission carried more than six times the risk of dying within a month compared with a similar patient whose BUN was in the lowest range.

The pattern extended well beyond the first month. Similar trends were observed for mortality at 90 days and at 365 days, indicating that the prognostic signal embedded in that single early measurement persists across both the acute crisis and the longer recovery period. Kaplan-Meier curves, which plot the proportion of patients surviving over time, showed significantly lower survival probabilities among patients with higher BUN levels, with a log-rank P value below 0.001, meaning the probability that such a pronounced separation between survival curves arose by chance is less than one in a thousand.

One of the most technically interesting aspects of the study emerged from restricted cubic spline analysis, a modeling technique that allows researchers to detect non-linear relationships between a continuous exposure and an outcome without imposing arbitrary assumptions about the shape of the curve. Rather than a simple straight-line relationship in which risk climbs uniformly with every increment of BUN, the spline analysis revealed a non-linear association with an inflection point at 51.3 milligrams per deciliter. This finding implies that the mortality risk associated with BUN does not increase at a constant rate; instead, the relationship changes character around this threshold, information that could prove valuable for clinicians attempting to translate the biomarker into a practical decision rule. The existence of such an inflection point suggests that risk stratification based on BUN may be most informative when the measurement is interpreted in relation to this turning point rather than as a purely linear scale.

Why should a measure of nitrogen waste in the blood carry such prognostic weight in myeloma patients? The biology offers several plausible explanations. BUN rises when the kidneys fail to clear urea, a byproduct of protein metabolism, and renal impairment is a hallmark complication of multiple myeloma. Malignant plasma cells produce excessive amounts of monoclonal immunoglobulin light chains that can precipitate in renal tubules, causing cast nephropathy, while hypercalcemia, dehydration, and infections common in advanced myeloma further compound kidney stress. Elevated BUN in an ICU patient with myeloma therefore functions as an integrated signal reflecting the convergence of tumor burden, renal damage, catabolic state, and volume depletion. It is precisely this composite nature that makes the biomarker so informative: a single number captures the physiological consequences of multiple pathological processes that each independently threaten survival.

The authors conclude that elevated BUN levels within 24 hours of first ICU admission are independently associated with short-term and long-term mortality in critically ill patients with multiple myeloma, and they propose that BUN may serve as a simple and readily available biomarker for early risk stratification in this population. In a plain language summary accompanying the paper, the researchers emphasize that BUN is a valuable indicator of disease outcomes and that their findings reveal elevated levels measured within the first day of intensive care are significantly associated with adverse outcomes. The practical appeal is obvious: unlike genomic profiling or specialized imaging, BUN is measured on routine chemistry panels in essentially every hospital in the world, requires no additional cost or delay, and can be incorporated into admission assessments immediately.

Several caveats temper the enthusiasm. As a retrospective cohort study drawing on a single-center database, the analysis establishes association rather than causation, and the findings require validation in external cohorts before BUN thresholds can be embedded in clinical guidelines. The study was funded by the Yunnan Clinical Research Center for Oral Diseases and the Yunnan Clinical Medical Center Research Project, and the authors declare no competing interests. Nevertheless, the work exemplifies a growing trend in critical care research: mining large electronic health record databases to rediscover the prognostic value of inexpensive, ubiquitous measurements that modern medicine has largely taken for granted. For patients with multiple myeloma fighting for their lives in intensive care units, and for the physicians deciding how aggressively to treat them, a number that has appeared on laboratory reports for decades may now carry life-or-death significance.

Subject of Research: Prognostic value of blood urea nitrogen for mortality in critically ill patients with multiple myeloma

Article Title: Association between blood urea nitrogen and prognosis in critical patients with multiple myeloma: a retrospective cohort study

Article References: Association between blood urea nitrogen and prognosis in critical patients with multiple myeloma: a retrospective cohort study. (n.d.). https://doi.org/10.1186/s12885-026-17200-x

Image Credits: AI Generated

DOI: 10.1186/s12885-026-17200-x

Keywords: blood urea nitrogen, multiple myeloma, MIMIC-IV, prognosis, biomarker, mortality, intensive care, retrospective cohort study, renal impairment, risk stratification, Kaplan-Meier analysis, Cox regression