Monocytopenia and Its Impact on Outcomes in Patients with Severe Sepsis Including Septic Shock: A Case-Control Study.
Ming Wu,Hao-li Li,Ying-yi Luan,Yang Feng,Yong-ming Yao
DOI: https://doi.org/10.1097/shk.0000000000001493
2019-01-01
Shock
Abstract:To the Editor: We read with great interest the review article recently published in Shock by Chung et al (1), entitled “Circulation monocyte counts and its impact on outcomes in patients with severe sepsis including septic shock”. The authors have demonstrated and highlighted that circulation monocyte counts influenced the outcome in patients with severe sepsis including septic shock, especially the 28-day mortality. Some points that tickled our interest to give a comment about this research article are incompleteness of whole story. First, this study was performed to evaluate the risk factors associated with the 28-day mortality, rate of bacteremia, and organ dysfunction in patients with severe sepsis including septic shock and further found monocyte counts were associated with mortality, the rate of bacteremia, and organ dysfunction in patients with sepsis. Just in terms of experimental design, it was a case-control study, but not a cohort study (just like authors statements in their limitations). Because case-control study is a research design used by researchers where the research begins with an outcome to comprehend the cause, the main concern in case-control studies is that they cannot predict causation, although they can present possible risk factors. The time sequence is from the back to the forward in case-control study and they are more prone to bias. A cohort refers to a large group of people who share similar characteristics. The researcher identifies those who have the disease and those who do not and study both groups with the exposure factors of identifying the risk factors. The specialty in cohort studies is that they can be longitudinal studies that go on for months or even years and the time sequence is from the front to the back. Second, there are many confounding factors effecting case and control groups. Pair group and individual match are used to decrease confounding. Selecting one or two parameters to match, and, 1:1–1:n matching is set in case and control. Although in this study, the numbers of the patients are no match between the case group and the control group (287/1725), no related parameters are matched, so there may be several confounding or selective bias, just like age and so on. Third, the patients in the groups are older (71.0 ± 13.3 versus 74.4 ± 12.6) than other study (2–4). Age is an independent risk factor for nonsurvivor in the elderly patients (hazard ratio 1.03 as shown in Table 4). Therefore, from the perspective of experimental design, age or (and) sex could be matched. Fourth, as shown in Table 4, Simultaneous multiple linear regression and Cox proportional hazard analysis were used, but there was little clinical significance for the outcome within 28 days because the time span was too short, maybe use anyone is OK. Fifth, in nonsurvivor group, high Sequential Organ Failure Assessment score, low albumin, and body temperature were all confounding factors, which affected the judgment of outcome. Finally, the mortality rate in Table 1 was 16.7%, whereas the mortality rate in Table 3 was 14.3%. It might be a clerical error. In addition, the mortality rate of patients in this study is lower than other international levels (4, 5), it may be the definition of sepsis or septic shock. The follow-up intervention that from March 2010 to June 2016 is great change because of the time span too long, and it is also a selective bias.