← Latest papers
📄 medicine

ICU day-1 arterial bicarbonate and 30-day mortality in adults with acute pancreatitis discharge diagnoses

This retrospective study of 1,280 adults with acute pancreatitis in the MIMIC-IV database found that while lower ICU day-1 arterial bicarbonate levels are significantly associated with higher 30-day mortality, this relationship largely reflects concurrent critical illness severity rather than serving as an independent, acute pancreatitis-specific prognostic marker.

Original authors: Yao Zheng

Published 2026-10-06
📖 5 min read🧠 Deep dive

Original authors: Yao Zheng

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

When the body's internal chemistry goes out of balance, the blood can become too acidic, a state that often signals a serious struggle within the body. One way doctors measure this balance is by looking at bicarbonate, a substance in the blood that acts like a sponge, soaking up excess acid to keep the body's environment stable. In the critical care unit, where patients are fighting for their lives, this number is checked frequently because it offers a quick snapshot of how well the body is handling stress. Acute pancreatitis, a sudden and painful inflammation of the pancreas, is a condition that can rapidly escalate from a manageable illness to a life-threatening emergency. While doctors have long known that severe cases often involve a drop in bicarbonate, it has remained unclear whether this drop is a specific warning sign unique to pancreatitis or simply a general indicator that a patient is critically ill from any cause. Understanding this distinction matters because if the low number is just a sign of general sickness, treating the pancreatitis specifically might not change the outcome, whereas if it is a unique signal, it could help doctors predict who needs the most urgent help.

Researchers set out to investigate this question by examining a vast collection of medical records from a large hospital database. They focused on adults who had been admitted to the intensive care unit with a diagnosis of acute pancreatitis. The team looked specifically at the very first measurement of arterial bicarbonate taken within the first day of the patient's stay in the intensive care unit. They then followed these patients to see who survived the first thirty days and who did not. The study included over 1,200 patients, a large group that allowed the researchers to look for patterns with a high degree of confidence. They compared the patients who had lower bicarbonate levels against those with higher levels, carefully accounting for other factors that could influence survival, such as the patient's age, existing health conditions, and the overall severity of their organ failure as measured by a standard scoring system used in intensive care.

The results showed a clear link between lower bicarbonate levels and a higher risk of death within thirty days. Among the patients whose bicarbonate levels were below 22 millimoles per liter, about 22.5 percent died within a month. In contrast, only 9.5 percent of those with levels at or above 22 millimoles per liter died in the same timeframe. Even after the researchers adjusted for how sick the patients were upon arrival, the association held true: for every five-unit drop in bicarbonate, the odds of dying increased. This suggests that a low bicarbonate level is indeed a marker of a grim prognosis for these patients. However, the study also revealed a crucial nuance. When the researchers adjusted their analysis to account for the severity of organ failure, the strength of the link weakened significantly. This indicates that the low bicarbonate level is likely not a unique warning sign specific to the pancreas itself, but rather a reflection of the body's overall struggle with critical illness. In other words, the low number tells doctors that the patient is very sick, but it does not necessarily tell them that the pancreatitis is the sole cause of that sickness.

The researchers also tested whether this finding held up when they looked only at patients who had confirmed enzyme levels in their blood, which provides stronger evidence that the diagnosis of pancreatitis was correct. In this smaller group, the link between low bicarbonate and death remained, though the numbers were less precise due to the smaller sample size. Conversely, when they looked at a group defined only by the hospital discharge code without enzyme confirmation, the link disappeared. This suggests that the finding is robust for patients who are clearly suffering from the disease, but it reinforces the idea that the bicarbonate level is a general barometer of critical illness rather than a specific tool for predicting the course of pancreatitis alone. The study did not find that adding bicarbonate to the blood would improve survival, nor did it suggest that this single number should replace existing methods for judging how sick a patient is. Instead, it confirms that a low bicarbonate level is a readily available, inexpensive signal that a patient with acute pancreatitis is facing a severe systemic crisis.

Ultimately, the paper concludes that while a low arterial bicarbonate level on the first day of an intensive care stay is a strong predictor of death, it functions more as a mirror of the patient's overall critical condition than as a specific crystal ball for pancreatitis. The drop in this chemical likely reflects a combination of poor blood flow, kidney stress, and the body's attempt to cope with widespread inflammation. For doctors, this means the number is valuable for recognizing the severity of the situation, but it should be interpreted alongside other signs of organ failure rather than viewed in isolation. The study provides a clearer picture of how the body's chemistry responds to this emergency, reminding us that in the intensive care unit, a single number often tells a story about the whole patient, not just one organ.

Drowning in papers in your field?

Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.

Try Digest →