Can Baseline Complete Blood Count Parameters Predict Incident Anaemia Among Patients with Rifampicin-Resistant Tuberculosis Receiving Linezolid-Based Regimens in Uganda? A Multicenter Retrospective Cohort Study

Tuberculosis remains one of the world's most persistent infectious diseases, but a specific form known as rifampicin-resistant tuberculosis poses a particularly difficult challenge because the bacteria do not respond to the standard first-line antibiotics. To fight this resistant strain, doctors rely on a powerful drug called linezolid, which has become a cornerstone of modern treatment regimens. However, like many potent medicines, linezolid comes with a cost. It can suppress the bone marrow, the spongy tissue inside bones where the body produces blood cells. When this suppression occurs, the body makes fewer red blood cells, leading to a condition called anaemia. This drop in red blood cells reduces the blood's ability to carry oxygen, often leaving patients feeling exhausted, weak, and at risk of needing blood transfusions or stopping their life-saving treatment altogether.

The critical question for doctors in resource-limited settings is how to predict who will suffer from this side effect before it happens. In many clinics, frequent blood testing is difficult to arrange, so physicians often rely on a single blood test taken before treatment begins. This test, known as a complete blood count, measures the levels of red blood cells, white blood cells, and platelets. The hope is that these initial numbers might act as a warning sign, revealing which patients are vulnerable to developing anaemia later, even if their blood counts look normal at the start. Understanding whether these early markers can forecast future trouble is essential for protecting patients who are already fighting a difficult battle.

In a recent study conducted across four hospitals in Uganda, researchers set out to answer this exact question. They looked back at the medical records of adults who were being treated for rifampicin-resistant tuberculosis with linezolid-based regimens between 2020 and 2024. To ensure they were studying new cases of anaemia rather than pre-existing conditions, the team focused only on patients who started treatment with healthy, normal levels of haemoglobin, the protein in red blood cells that carries oxygen. The researchers tracked these patients over time, watching to see who would develop anaemia during their treatment and how quickly it would happen.

The results revealed that the risk is significant and often appears sooner than expected. Among the 101 patients who started with normal blood counts, nearly half developed anaemia at some point during their treatment. This means that having a normal blood test before starting the drug does not guarantee safety. The timing of these events was striking: more than two-thirds of the new anaemia cases occurred within the first three months of treatment. In fact, the median time to the first drop in blood levels was just two months. This rapid onset suggests that the body's reaction to the medication can be swift, catching patients off guard if they are not monitored closely during the early stages of therapy.

When the researchers dug deeper to find out which patients were most at risk, they found that the initial haemoglobin level was a powerful predictor. Patients who started with haemoglobin levels closer to the lower end of the normal range were significantly more likely to develop anaemia later on. For every small increase in the starting haemoglobin level, the risk of developing the condition dropped substantially. Sex also played a clear role; men were less likely to develop anaemia than women, even when starting with similar blood counts. This difference likely reflects the fact that men naturally have higher haemoglobin levels, giving them a larger buffer before dipping into the anaemic range, though other biological factors may also be involved.

Interestingly, the study found that other common blood markers did not offer the same warning. The initial counts of white blood cells, which fight infection, and platelets, which help blood clot, showed no independent link to the development of anaemia. This is a crucial finding because it tells doctors that a normal white blood cell or platelet count does not mean a patient is safe from red blood cell problems. The only reliable early indicators were the starting haemoglobin level and the patient's sex.

The study also highlighted the severity of the situation. While most cases of anaemia were manageable, a small number of patients experienced severe drops in haemoglobin that required urgent attention. The fact that so many cases appeared within the first three months suggests that the most critical window for monitoring is right at the beginning of treatment. For doctors in Uganda and similar settings, this means that patients starting with lower-normal haemoglobin levels should be watched more closely during these early months. By identifying these vulnerable individuals early, medical teams can intervene before the condition becomes severe, ensuring that patients can continue their tuberculosis treatment without the added burden of dangerous anaemia. The findings do not suggest stopping the use of linezolid, which remains a vital tool, but rather point toward a smarter, more targeted way of using the information already available in a simple blood test to protect patient safety.