In this study, we demonstrated that DNI, which reflects the number of circulating granulocyte precursors in the blood, correlated with the severity of sepsis in critically ill patients admitted to the medical ICU. The elevation of DNI value preceded the onset of organ/circulatory failure, thus contributing to identifying patients with an impending risk of developing severe sepsis/septic shock.
In the present study, increased DNI values at the time of ICU admission were significantly associated with the presence of severe sepsis/septic shock and overt DIC. These results are consistent with a previous report by Nahm et al. who concluded that DNI was closely related to the presence of overt DIC, bacterial isolation rate, and mortality in patients with suspected sepsis . In another study by Ansari-Lari et al., the percentage of immature granulocytes correlated better with infection and positive blood culture results than the WBC count , but the sensitivity was low (40% sensitivity at 90% specificity). Consequently, the authors suggested that high cut-off levels for the percentage of immature granulocytes might be required to reliably predict infection or positive blood culture results. In that study, however, the authors tested the usefulness of immature granulocytes as a predictor of infection, not sepsis. In contrast, we evaluated the clinical usefulness of DNI in sepsis (including severe sepsis/septic shock). These different inclusion criteria may have yielded different sensitivity, specificity, and even higher optimal cut-off value for DNI in our results.
Severe forms of sepsis are associated with DIC, and DIC is often present before the onset of sepsis . We found that DNI correlated with SAPS 3, SOFA, and DIC score. Similar to our results, a previous report by Nahm et al. also demonstrated a significant relationship between DNI and DIC-related parameters, including platelet count, PT, aPTT, and antithrombin III . These findings suggest that DNI may be linked to a hypercoagulable state which is associated with sepsis, and that DNI may reflect the clinical severity of critically ill patients with sepsis.
In our results, DNI values were higher in the severe sepsis/septic shock group compared to the sepsis group. In line with this finding, the Cochrane-Armitage trend test also showed that the proportion of patients with overt DIC and severe sepsis/septic shock gradually increased with the increase in DNI values. Importantly, the proportion abruptly increased when DNI was higher than 4.3% (the second quartile value of DNI). In our ROC analysis, the diagnostic value of DNI for severe sepsis/septic shock was superior to WBC, ANC, or other widely available laboratory markers. The optimal cut-off value of DNI for predicting severe sepsis/septic shock was 6.5%. Taken together, our data suggest that careful attention may be required in patients with suspected infection for possible concomitant DIC and/or severe sepsis/septic shock if DNI value increases up to 4-6% or more.
A recent study showed that mortality was correlated to the duration of hypotension before the start of antibiotic treatment . Therefore, it is very important role for clinicians to identify patients who are at risk of developing severe sepsis/septic shock before the signs of organ dysfunction or circulatory failure appear. In the present study, DNI values had already increased before the onset of organ/circulatory failure in 82% of the patients with severe sepsis/septic shock, suggesting that DNI may help to identify patients with an imminent risk of developing severe sepsis/septic shock. Given that the process of granular leukocyte differentiation starts from immature granulocyte formation, the change in DNI may have preceded the change in absolute numbers of WBC or neutrophil, thus contributing to predicting the development of severe sepsis/septic shock. Based on our data, we suggest that the finding of an increased DNI value should alert clinicians to start fluid resuscitation or to change antibiotic treatment.
Several limitations of our study should be mentioned. First, this was a single-center study and the sample size was relatively small. Second, the elevation of immature granulocytes is not specific for infection and may be observed in various other conditions, including myeloproliferative disorders, chronic inflammatory diseases, tissue damage, acute hemorrhage, and neoplasia . Because DNI is also one of the leukocyte-related parameters, there may be a lack of sensitivity or specificity for DNI as a severity marker of sepsis in this group. Third, we did not evaluate the comparative advantage of using DNI over procalcitonin which may have a role in reducing antibiotic exposure of critically ill patients  and serve as a useful complementary comparator for prediction of survival outcome in postoperative patients with severe sepsis . Because the main focus of the present study was DNI, and the decision to check procalcitonin level was left to the attending physicians, the data for procalcitonin were obtained only from 63 (61%) of 103 patients, thereby yielding relatively lower discriminative power (AUC 0.65 [95% CI, 0.54-0.77], data not shown) compared to the previous reports . More studies with large number of patients are required to validate the clinical usefulness of DNI as a severity and prediction marker of sepsis. Further studies are also warranted to investigate the additional benefit of combining DNI with other biomarkers, such as procalcitonin, to improve their predictive power.