Xianshi Zhou

Human leukocyte antigen-DR expression might predict outcomes in severe sepsis, but diabetes mellitus cannot be ignored

We read with great interest, but also with some concern, the paper by Drewry et al. recently published in Critical Care [1]. In this observational study, the authors concluded that human leukocyte antigen-DR (HLA-DR) expression might be a more accurate predictor of mortality and acquisition of secondary infections than lipopolysaccharide-stimulated TNF-α production in critically ill patients. The primary concern with the article is that it overlooks the effects that diabetes might exert on the outcomes. Notably, the ratio of comorbid diabetes was statistically higher in the survivors than the nonsurvivors (P = 0.034). Diabetes has close and complex relationship with sepsis—it can not only alter the immune response (including HLA-DR expression) [2, 3], but also influence the mortality rate and risk of infection in patients with sepsis [2]. In addition, hyperglycemia, insulin [2], and obesity [4] secondary to diabetes can also impact on both the immune system and critical outcomes. Thus, it did not seem persuasive that HLA-DR expression could completely predict outcomes without adjusting for the covariate of diabetes. Similarly, the APACHE II and SOFA scores, which reflect the severity of the disease, were statistically lower in the survivors than the non-survivors (P < 0.001 and P = 0.009, respectively). However, in a prospective study on mHLA-DR [5], in which the severity level based on SAPS II or SOFA score was adjusted, the original statistical relationship between early mHLA-DR downregulation and outcomes in the whole population disappeared (P > 0.05). Thus, it is plausible and necessary to adjust APACHE II and SOFA scores for statistical analysis in this paper. The two inaccuracies described above are examples of colinearity, which suggest that not only univariate but also multivariable analysis should be carried out in this study to exclude interactions between variables and make the conclusion more prudent and accurate. In addition, we do not understand why the APACHE II and SOFA scores were calculated without the neurological component, which can be easily done as the study was described as prospective. Moreover, we suggest that receiver operating characteristics (ROC) curves for the outcomes and these two markers be drawn to compare and determine which is the better predictive marker and what is the best, most pragmatic threshold (i.e., maximized sum of sensitivity and specificity). We appreciate the meaningful work of Drewry et al., which provides us with a new perspective on predictive indicators in severe sepsis, though the statistical methods in the study need some amendments.

Some questions on the use of Xuebijing in treating patients with sepsis

A meta-analysis of randomized controlled trials on evaluating the clinical efficacy of Xuebijing in treating patients with sepsis had been recently published in this journal, however, in our opinion, at least 6 questions or doubts existed about the content and methodology of the systematic review, thus, we write this letter to the journal to share these questions with the editor and readers with the hope of getting these questions solved and learning more about the true effects which Xuebijing should have on in treating patients with sepsis.

Acute appendicitis caused by Schistosoma japonicum

We read with interest the case of a 74-year-old woman with cute appendicitis caused by Schistosoma japonicum reported by aiti et al. [1]. And we have some comments on this report. First, the article mentioned that ‘schistosomiasis causes around 00,000 deaths per year.’ However, we do not agree with this view. s an endemic communicable disease caused by blood flukes of the enus Schistosoma, schistosomiasis infection is found in more than 50 million people, with almost 800 million at risk worldwide [2]. here are estimated up to 280,000 deaths annually related to schisosomiasis just in sub-Saharan Africa in the early 2000s [3,4]. Also he number of deaths varies as different prevention and control trategies being applied and influenced by the climate, ecosysem and economic capability, and etc. So the figure provided by he authors may not be quite exact. Actually, there are six schisosome species known to infect man (S. japonicum, Schistosoma ansoni, Schistosoma hematobium, Schistosoma intercalatum, Schisosoma mekongi, and Schistosoma malayensis) [5] not just ‘Five’ [1]. Second, the authors put forward an important issue of ‘neglected ropical endemic disease’ which deserves to be concerned and esolved. With the ease of international travel and frequent internaional communications, more and more neglected tropical endemic iseases had been poured into non-endemic developed counries from endemic developing countries [5]. TropNetEurope had eported more than 800 imported schistosomiasis cases in 2003, ost of which were immigrants and refugees, and European ourists occupied only 1/7. So schistosomiasis is arising in Western ountries and may not be a unusual aetiology for acute appendicitis or the future [6]. The last but not least, as a country struggling with schistoomiasis for thousands of years, China has accumulated a lot of uccessful experience in S. japonicum control over the past decades. hinese medical literature has covered plenty of cases of acute ppendicitis caused by S. japonicum. Compared with other others ypes of appendicitis, appendicitis caused by S. japonicum basely xists these relatively specific characters: (1) most patients are pyrexia, less can be hyperpyrexia, while others appendicitis often as a hyperthermia more than 39 ◦C. (2) Metastatic right lower uadrant pain, the metastatic time is greatly shortened, mostly resent a fixed or reiterant right lower quadrant pain accompanied ith nausea, emesis, obvious tenderness and rebound tenderess in that region. (3) Other abdominal sign is mild or absent. 4) Usually elevated eosinophil count, normal or mildly elevated hite blood count (10–15 × 109/L), while others types usually high

Clinical study of a new Modified Early Warning System scoring system, some lingering doubts

The clinical study byQin et al. [1] in a recent issue of Journal of Critical Carewas fascinating andmeaningful, we read it with great interest. The authors concluded that the new Modified Early Warning System (MEWS) scoring system could be an independent indicator to reflect shock severity and it had higher predictive efficacy in septic shock, especially for 28-day prognosis. Although the study sounds scientific,we still have some lingering doubts on this research. The study should be prospective as it claimed that written informed consent was obtained from all patients which carried out between January 2013 and January 2014. Nevertheless, the referred 2 criteria [2,3] for cardiogenic shock in the study both published after January 2013(the online versions came out on September 5, 2013 [3] and February 6, 2014 [2]), how could it happen?What ismore, one cited reference [2] for the criteria of cardiogenic shockwas just a letterwhere no criteria mentioned, thus the authors had better explain the reasons. The authors claimed that the criterion of septic shockwas the detection of serious infection and sepsis according to the 2008 SCC (Surviving Sepsis Campaign) [4], however, the criterion for septic shock was not so simple in the guideline, Actually, sepsis induced hypotension was defined as a systolic blood pressure (SBP) of b90 mmHg or mean arterial pressure b 70 mmHg or a SBP decrease N40 mmHg or b2 standard deviation below normal for age in the absence of other causes of hypotension. And septic shock was defined as sepsis induced hypotension persisting despite adequate fluid resuscitation in the guide [4]. The data in the article perhaps need rechecking again. As shown in Table 1, serum potassium concentration was (25.26 ± 49.12) and (3.86 ± 0.97) in the non-surviving group and surviving group respectively. Since the normal distribution of serum potassium concentration is from 3.5 mmol/L to 5.5 mmol/L and serum potassium concentration ≧7.0mmol/L is an indication for urgent hemodialysis that it was incredible to believe themean serumpotassiumconcentration in the deceased

Some questions on Shenfu injection for patients with sepsis

☆ Funding: YY was supported by a grant from Guang Science and Technology (grant number: 2015A02021009 role in the preparation, drafting, review, or approval of t submit the manuscript for publication. ☆☆ Authors contributions: XZ and YY wrote the main tex ceived the ideas central to the commentary. All authors re ⁎ Corresponding author at: Department of Emergency, of Chinese Medicine, 111 Dade Road, Yuexiu District, Gua E-mail addresses: 13660638204@163.com (X. Zhou), 1183894950@qq.com (B. Zhou), 1404387022@qq.com (G 1 Xianshi Zhou and Ye Ye contributed to thework equal first authors.