By Richard R. Watkins, MD, MS, FACP, FIDSA, FISAC

Professor of Internal Medicine, Northeast Ohio Medical University; Division of Infectious Diseases, Cleveland Clinic Akron General, Akron, OH

Dr. Watkins reports no financial relationships relevant to this field of study.

SYNOPSIS: A single-center, prospective, observational study found that an elevated procalcitonin level was associated with an increased risk of death in patients with pulmonary tuberculosis.

SOURCE: Osawa T, Watanabe M, Morimoto K, et al. Serum procalcitonin levels predict mortality risk in patients with pulmonary tuberculosis: A single-center prospective observational study. J Infect Dis 2020;222:1651-1654.

Tuberculosis (TB) remains the leading cause of death due to infection worldwide. Yet, little is known about the risk of mortality in patients before and after the initiation of anti-tuberculosis therapy. Osawa and colleagues sought to determine the utility of serum procalcitonin (PCT) in identifying patients with TB at an increased risk of death.

The investigators conducted a single-center, prospective observational study that enrolled patients between December 2013 and December 2015. Patients were included who were at least 18 years of age, were newly diagnosed with active pulmonary TB, and were not receiving TB treatment. The diagnosis of pulmonary TB was in patients with a new radiographic pulmonary infiltrate at hospitalization, positive smear and/or Mycobacterium tuberculosis polymerase chain reaction (PCR) test of sputum or gastric juice, and subsequent positive M. tuberculosis culture. Patients were excluded if they had disseminated TN (defined as positive M. tuberculosis PCR or culture of blood, urine, or cerebral spinal fluid) or were treated for TB in the preceding 10 years. PCT levels, other blood tests, and chest X-rays were obtained on admission (day 0) and on days 7, 14, and 28 after admission. Anti-TB therapy was started within two days of admission.

There were 252 patients enrolled in the study, including 213 survivors and 39 non-survivors, none of whom were human immunodeficiency virus (HIV)-positive. The two groups were well balanced in terms of coexisting medical conditions, use of immunosuppressive agents, and presence of cavitary lung lesions. However, the median age for survivors was 65 years, compared to 86 years for non-survivors (P < 0.001). More non-survivors had acute respiratory failure (n = 19, 48.7%) and bilateral lesions (n = 35, 89.7%) compared to survivors (n = 12, 5.5% and n = 117, 54.9%, respectively).

PCT levels were greater at all four time points for the non-survivors compared to the survivors. The Youden index calculations that predicted patient death were 0.13 on day 0, 0.05 on day 7, 0.12 on day 14, and 0.06 on day 28, with odds ratios (95% confidence intervals [CI]) of 7.9 (95% CI, 3.2-19.7), 14.3 (95% CI, 3.1-66.1), 20.0 (95% CI, 5.7-69.6), and 7.3 (95% CI, 1.8-29.5), respectively. PCT levels decreased over the course of treatment, but the decreases were not statistically significant between the survivors and non-survivors. Finally, PCT levels were weakly associated with sputum acid-fast load on day 0, but not at other time points (P = 0.08).


This study found that serum PCT levels predicted mortality in patients treated for pulmonary TB. The investigators calculated a priori the cohort size needed to determine whether PCT is an accurate mortality risk indicator. Thus, PCT theoretically could be useful as both a marker for pretreatment predictions and for posttreatment monitoring of mortality in patients with pulmonary TB.

Another interesting finding was that anti-TB therapy did not cause significant reduction in PCT levels in either survivors or non-survivors. This may have been a reflection of non-tubercular inflammation.

As the authors mentioned, certain underlying clinical conditions, such as cancer, chronic obstructive pulmonary disease, coronary artery disease, liver disease, and renal failure, are known to be risk factors for TB-associated mortality. Perhaps inflammation caused by these conditions leads to increased mortality, and additional investigation of this hypothesis seems warranted. Furthermore, if this hypothesis is true, then agents that modulate inflammation (e.g., steroids, nonsteroidal anti-inflammatory drugs, hydroxychloroquine, and N-acetylcysteine) might be efficacious adjuncts to anti-TB medications.

The study had some limitations. First, since it was a single-center study conducted in Japan, the results might not be generalizable to other populations and geographic areas. Second, the non-survivors were significantly older than the survivors (median age 86 years vs. 65 years, respectively). It would have been interesting if the authors had determined whether there was any association between age and PCT levels in survivors vs. non-survivors. Third, a standard course of therapy for pulmonary TB is six months, yet the authors followed PCT levels only up to 28 days after admission. Further investigation is needed to measure PCT levels over the entire course of therapy. Finally, only cases of active pulmonary TB were included. Whether PCT is elevated in cases of latent TB is unknown and also warrants further study.

Previous studies have demonstrated that PCT can predict mortality in patients with sepsis and pneumonia. Osawa and colleagues have provided preliminary evidence that PCT also may have a similar role in pulmonary TB, although further research is necessary before routine PCT testing in this population can be recommended. Whether PCT has a role in other mycobacterial infections (e.g., Mycobacterium avium complex [MAC]) is another topic for investigation as well.