Secondary Logo

Medicine Correspondence Blog

The Medicine Correspondence Blog allows authors to post Letters to the Editors, Reviews, and other editorial writings that are not considered original research.

Thursday, November 8, 2018

With interest we read the systematic review of Song et al1, aiming to investigate the clinical outcomes of early enteral nutrition (<48 hours) in severe acute pancreatitis. We would like to address two methodological issues that may have influenced outcomes.

The authors recognize that enteral nutrition may have beneficial effects on the intestinal barrier function. Surprisingly, however, the authors created a control-group which included studies on enteral and studies on parenteral nutrition. The favorable results of early enteral nutrition in this meta-analysis could have very well occurred due to the detrimental effects of the parenteral nutrition in the control group.

Furthermore, in the meta-analysis focusing on the outcomes of systemic infections and gastro-intestinal symptoms, the authors have added up all the different events, as reported in the original article, not taking into account that these events could have occurred concurrently (i.e. multiple events in a single patient). This will introduce bias and may lead to overestimation of the effect of the intervention studies in the meta-analysis.

By preservation of the gut barrier function and reduction of bacterial translocation, enteral nutrition has the potential benefit of reducing complications in acute pancreatitis. Therefore, the use of parenteral nutrition should be discouraged, and only be used if enteral nutrition is not possible.

Therefore, we would like to emphasize the conclusions of the only multicenter randomized controlled trial that truly compared early enteral feeding (<24h) vs delayed enteral feeding (on demand >72h): enteral tube feeding can be limited to patients with predicted severe pancreatitis who have inadequate intake after 72 hours after onset of symptoms. Initiating enteral feeding within 24 hours of start of symptoms did not improve outcomes2


Author Correspondence

Sven M. van Dijkm, Department of Surgery, Amsterdam Gastroenterology and Metabolism, Amsterdam UMC, University of Amsterdam and Department of Research & Development, St. Antonius Hospital, Nieuwegein

Olaf J. Bakker, Department of Surgery, St. Antonius Hospital, Nieuwegein

Hjalmar C. van Santvoort, Department of Surgery, St. Antonius Hospital, Nieuwegein and Department of Surgery, University Medical Center Utrecht, Utrecht

Marc G. Besselink, Department of Surgery, Amsterdam Gastroenterology and Metabolism, Amsterdam UMC, University of Amsterdam


References

[1] Song J, Zhong Y, Lu X, et al. Enteral nutrition provided within 48 hours after admission in severe acute pancreatitis: a systematic review and meta-analysis. Medicine. 97(34):e11871, August 2018.

[2] Bakker OJ, van Brunschot S, van Santvoort HC, et al. Early versus on-demand nasoenteric tube feeding in acute pancreatitis. N Engl J Med 2014;371:1983-1993


Friday, October 26, 2018

We read the published study by Jialing Shi and Liying Wei that effect of combined parenteral and enteral nutrition versus enteral nutrition alone for critically ill patients[1],carefully.However,we found some points worth discussing in the paper.

Firstly,according to requirements of PRISMA[2],detailed retrieval strategies for at least one database are provided in the article, and it is also easy for others to repeat experiments.However, the author did not mention it in the paper.

Secondly,the heterogeneity of meta analysis was tested by I2 in the paper.But the author said in the article that the x2 test was used to quantify statistical heterogeneity. So this is an obvious mistake.In addition, we think the author's definition of the scope of heterogeneity is not accurate enough.It is generally believed in meta analysis that I2<25% was considered as low-level heterogeneity, 25% to 75% as moderate-level, and >75% as high-level heterogeneity[3].In this paper,the source of heterogeneity has not been explored.So we think it's not rigorous.Heterogeneity sould be explained by subgroup analysis or meta-regression.

Thirdly,in the meta-analysis, the author used the random effect model without distinction.So we don't think the conclusion is accurate enough.The use of fixed effect model and random effect model depends on heterogeneity in the meta-analysis.If there was no heterogeneity, fixed effect model can be used. When substantial Heterogeneity observed, random effect model was utilized.

Fourthly,Publication bias will affect the final results of meta analysis[4], so the recognition and processing of publication bias is an important step in systematic evaluation.However, publication bias was not detected in this meta-analysis.Therefore, we cannot evaluate whether the final conclusion of this meta-analysis is affected by publication bias.

In summary,meta-analysis is an important means of evidence-based medicine.It is necessary to carry out meta-analysis according to PRISMA's requirements.And it can guarantee the scientificity and rigor of meta analysis.


Author Correspondence

Pengfei Sun, Department of Medicine, Qingdao University, China 

Yanjin Wang, Department of Plastic Surgery, Affiliated Hospital of Qingdao University, China 

Zhenyu Chen, Department of Plastic Surgery, Affiliated Hospital of Qingdao University, China 

Email: 18678330029@163.com

Tel: +86 18678330029


References

[1] Shi J, Wei L, Huang R, et al. Effect of combined parenteral and enteral nutrition versus enteral nutrition alone for critically ill patients: A systematic review and meta-analysis. Medicine (Baltimore) 2018;97(41):e11874.

[2] Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ 2009; 339: b2700.

[3] Huedo-Medina TB, Sánchez-Meca J, Marín-Martínez F, et al. Assessing heterogeneity in meta-analysis: Q statistic or I2 index. Psychol Methods 2006; 11(2): 193-206.

[4] Peters JL, Sutton AJ, Jones DR, et al. Comparison of two methods to detect publication bias in meta-analysis. JAMA 2006; 295(6): 676-680.

Thursday, September 20, 2018

With great interest we read the article by Jang et al.1 They showed that 8/20 (40%) heart valves of patients with culture-negative infectious endocarditis (IE) were Coxiella burnetii PCR-positive. None of these patients received adequate treatment and their mortality was higher than in C. burnetii PCR-negative patients.

 

This study describes a high percentage of missed diagnoses of Q fever endocarditis with a high mortality rate and hopefully raises awareness among medical specialists in South Korea. All medical specialists that care for patients with IE should be educated on how to recognize and treat Q fever endocarditis. When culture-negative IE is suspected or diagnosed, serological testing for C. burnetii is indicated in line with the modified Duke criteria.2,3 It surprised us that testing for C. burnetii serology had not been performed in these patients with culture-negative IE reportedly because of "low clinical suspicion" due to the low incidence of Q fever endocarditis.

 

Q fever endocarditis can present as a chronic low-grade infection with few signs or symptoms, but can cause life-threatening complications.4,5 When C. burnetii PCR is positive on blood or tissue in patients with suspected or diagnosed IE, the diagnosis of Q fever endocarditis is considered proven. This criterion is described in the Dutch consensus guideline on chronic Q fever6 as well as in the French guidelines from Eldin et al.7 and also applies to countries with low incidence of Q fever endocarditis.

 

Even if the diagnosis of Q fever endocarditis is made in retrospect for research purposes, a positive C. burnetii PCR on heart valves should never be discarded and treatment should have been initiated. We were surprised that only two C. burnetii PCR-positive patients were serologically tested for C. burnetii after their heart valves were C. burnetii PCR-positive and none of them were treated for Q fever endocarditis. Without adequate therapy, the higher mortality rate in the C. burnetii PCR-positive patients is not surprising at all. Therefore, we strongly advise to inform C. burnetii PCR-positive patients about their diagnosis, examine them for ongoing endocarditis and start treatment.

 

To conclude that C. burnetii PCR on excised heart valves may increase the diagnostic yield and reduce the number of missed cases is important, but it should not result in waiting until IE progresses as far as needing heart valve replacement before diagnostic testing for C. burnetii is performed. We recommend serological testing for C. burnetii in all culture-negative IE patients, in line with the modified Duke criteria.2 This way, adequate treatment can be started early in the course of the disease and surgery may not be necessary.


Patients with (suspected) Q fever endocarditis should preferably be referred to a center with experience in this field or, if this is not possible, an expertise center should be consulted. The data presented by Jang et al. clearly indicate the necessity to consider Q fever endocarditis in all cases of culture-negative IE and will hopefully increase awareness in South Korea for this disease.  


Author Correspondence

 S.B. Buijs, MD 

Email: S.B.Buijs-2@umcutrecht.nl  


References

[1] Jang YR, Song JS, Jin CE, et al. Molecular detection of Coxiella burnetii in heart valve tissue from patients with culture-negative infective endocarditis. Medicine (Baltimore). 2018;97(34):e11881.

[2] Li JS, Sexton DJ, Mick N, et al. Proposed modifications to the Duke criteria for diagnosis of infective endocarditis. CID. 2000;30:633-638.

[3] Habib G, Lancellotti P, Antunes MJ, et al. 2015 ESC Guidelines for the management of infective endocarditis: The Task Force for the Management of Infective Endocarditis of the European Society of Cardiology (ESC). Endorsed by: European Association for Cardio-Thoracic Surgery (EACTS), the European Association of Nuclear Medicine (EANM). Eur Heart J. 2015;36(44):3075-3128.

[4] Kampschreur LM, Delsing CE, Groenwold RH, et al. Chronic Q fever in the Netherlands 5 years after the start of the Q fever epidemic: results from the Dutch chronic Q fever database. J Clin Microbiol. 2014;52(5):1637-1643.

[5] ​van Roeden SE, Wever PC, Kampschreur LM, et al. Chronic Q fever-related complications and mortality: data from a nationwide cohort. ECCMID 2017 eLibrary Material of the 27th European Congress of Clinical Microbiology and Infectious Diseases; 2017; Vienna, Austria.

[6] Wegdam-Blans MC, Kampschreur LM, Delsing CE, et al. Chronic Q fever: review of the literature and a proposal of new diagnostic criteria. J Infect. 2012;64(3):247-259.

[7] Eldin C, Melenotte C, Mediannikov O, et al. From Q Fever to Coxiella burnetii Infection: a Paradigm Change. Clin Microbiol Rev. 2017;30(1):115-190.

 


Thursday, September 13, 2018

I read with interest the article by Peng et al.[1] entitled: "The long-term efficacy of STN vs GPi deep brain stimulation for Parkinson disease: A meta-analysis". However, this meta-analysis includes several methodological mistakes that worth attention.

In systematic reviews and Meta-analyses, the authors should exclude duplicate reports and select unique studies for inclusion in the evidence synthesis process. This is a common practice in this journal "Medicine" and following the guidelines of Cochrane Handbook of Systematic Reviews and Meta-analysis of interventional studies and the preferred reporting items of systematic review and meta-analysis (PRISMA statement) [2, 3].

In this meta-analysis, the authors included five articles. However, three of these five articles described the same patients; Follet et al.[4], Weaver et al.[5], and Katz et al.[6], are three reports from the same study, Veterans Administration Cooperative Studies Program #468 multicentre study. Because of pooling three reports from the same RCT, the impact of this multicentre RCT in the analysis was augmented and has led to incorrect results.

For example, in Figure 5, the authors pooled the three duplicate reports from the Veterans Administration Cooperative Studies Program #468 multicentre study (ClinicalTrials.gov numbers, NCT00056563 and NCT01076452.). According to Figure 5, the differences in PDQ ADL in the 3 studies were as follows: Follet 2010 (MD -4.5 with 95% CI from -9.11 to 0.11), Katz 2015 (MD -1.6 with 95% CI from -6.63 to 3.43), and Weaver 2012 (MD -4.0 with 95% CI from -10.39 to 2.39). It deserves our notice that the MDs of the three reports were not statistically significant. However, when the three duplicate studies are pooled in the meta-analysis model, the pooled MD was statistically significant (MD -3.36 with 95% CI from -6.36 to -0.36, P=0.03).

Secondly, Rodriguez-Oroz et al.[7] published a long-term report from a non-randomized study (DBS study group 2000 [8]). There is a methodological concern in pooling randomized and non-randomized studies in meta-analysis. However, this is not the main problem; The authors assessed the risk of bias in Rodriguez-Oroz et al.[7] using the Cochrane ROB tool, a tool for evaluating the risk of bias in RCTs[2]. Alternatively, the authors should have assessed the Rodriguez-Oroz et al.[7] study using the ROBINS-I checklist for non-randomized studies [9]. A more serious mistake here is that the authors marked the allocation concealment as (yes +) and the blinding as (yes +) which seems impossible since the Rodriguez-Oroz et al.[7] study was not RCT and is no concealed allocation or blinding in non-RCT studies.

Another issue in the risk of bias assessment is that the study of Weaver et al.[5] was marked as (no -) in terms of the allocation concealment and (unclear ?) in terms of other source of bias while Follet 2010 [4] and Katz 2015 [6] were considered as (yes +) and (no -) in terms of allocation concealment and other bias, respectively. Given that the three reports are describing the same multicentre RCT (Veterans Administration Cooperative Studies Program #468), the risk of bias in these three reports should be similar as they represent the same clinical trial. However, the authors mistakenly judged the risk of bias scores of Weaver et al.[5] (4 out of 6) while both Follet 2010 [4] and Katz 2015 [6] scored 6.

(3) It is known that higher scores of UPDRS indicate worse PD symptoms while a decrease in UPDRS-III indicates the improvement in motor functions. Therefore, the mean change in UP​DRS scores in all included studies was calculated as MD=post–baseline, therefore, the changes in UPDRS scores after DBS include a minus sign highlighting a decrease in the UPDRS score. Unfortunately, the authors ignored the direction of the scale and therefore, the presented effect estimates of individual studies and the pooled effect estimate are in the opposite direction.

It is important to highlight that this meta-analysis was based on several methodological errors and if corrected by removing the duplicate studies, most of the provided plots will not be reliable statistically due to an insufficient number of included studies (one study in some outcomes). The evidence provided by this meta-analysis is not robust and is not conclusive; Further studies are still needed to evaluate the long-term efficacy of STN DBS and GPi DBS in patients with advanced PD.


Author Correspondence

Ahmed Negida; Faculty of Medicine, Zagazig University, Zagazig, El-Sharkia, Egypt

Email: ahmed01251@medicine.zu.edu.eg


References

[1] Peng L, Fu J, Ming Y, et al. (2018) The long-term efficacy of STN vs GPi deep brain stimulation for Parkinson disease. Medicine (Baltimore) 97:e12153 . doi: 10.1097/MD.0000000000012153

[2] Higgins J, Green S (2011) Cochrane handbook for systematic reviews of interventions Version 5.1. 0. [updated March 2011]

[3] Moher D, Liberati A, Tetzlaff J, Altman DG (2009) Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 6:e1000097 . doi: 10.1371/journal.pmed.1000097

[4] Follett KA, Weaver FM, Stern M, et al (2010) Pallidal versus Subthalamic Deep-Brain Stimulation for Parkinson's Disease. N Engl J Med 362:2077–2091 . doi: 10.1056/NEJMoa0907083

[5] Weaver FM, Follett KA, Stern M, et al (2012) Randomized trial of deep brain stimulation for Parkinson disease: thirty-six-month outcomes. Neurology 79:55–65 . doi: 10.1212/WNL.0b013e31825dcdc1

[6] Katz M, Luciano MS, Carlson K, et al (2015) Differential effects of deep brain stimulation target on motor subtypes in Parkinson's disease. Ann Neurol 77:710–719 . doi: 10.1002/ana.24374

[7] Rodriguez-Oroz MC, Obeso JA, Lang AE, et al (2005) Bilateral deep brain stimulation in Parkinson's disease: a multicentre study with 4 years follow-up. Brain 128:2240–9 . doi: 10.1093/brain/awh571

[8] Deep-Brain Stimulation for Parkinson's Disease Study Group, Obeso JA, Olanow CW, et al (2001) Deep-brain stimulation of the subthalamic nucleus or the pars interna of the globus pallidus in Parkinson's disease. N Engl J Med 345:956–63 . doi: 10.1056/NEJMoa000827

[9] Sterne JA, Hernán MA, Reeves BC, et al (2016) ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. BMJ i4919 . doi: 10.1136/bmj.i4919​

Friday, September 7, 2018

We have read with interest the paper by Zhu et al. on the management of a patient with testicular Leydig cell tumor.1

The Authors reported a patient with testicular neoplasm that histological and immunohistochemical analyses revealed to be a Leydig cell tumor. As stated by the Authors, the immunohistochemical analysis showed that the tumor cells stained positively for inhibin and negatively for pan-cytokeratin, cytokeratins 8/18, alpha-feto ptotein, human corionic gonadotropin, CD 30, CD 99 and S-100.

The Authors throughout the paper report the term inhibin but they should had reported inhibin B that is mainly produced within the testis.  To this regard it is well known that inhibin B is the main protein physiologically secreted by Sertoli cells located within the seminiferous tubule under the control of FSH.2,3 Leydig cells do not secrete inhibin B but produce other proteins such as INSL3, that has been utilized to characterize Leydig cells an also Leydig cells tumors by immunohistochemical analysis.4

In the case report of the Authors, if the immunohistochemical analysis revealed the expression of inhibin (possibly B), it could be the case of a mixed Sertoli-Leydig cell tumor, although this kind of tumor is exceptionally rare in the male being more frequent in the ovary.5

 

Author Correspondence:

Marco Rossato, MD, PhD

University of Padova, Padova, Italy.

marco.rossato@unipd.it

 

Luca De Zorzi, MD,

Istituto Oncologico Veneto, Treviso, Italy

dezorziluca@gmail.com

 

 

References

[1] Zhu JLuan YLi H. Management of testicular Leydig cell tumor: A case report. Medicine (Baltimore) 2018;97:25.

[2] de Kretser DM, Robertson DM. The isolation and physiology of inhibin and related proteins. Biol Reprod 1989;40:33-47.

[3] Foresta C, Bettella A, Petraglia F, Pistorello M, Luisi S, Rossato M. Inhibin B levels in azoospermic subjects with cytologically characterized testicular pathology. Clin Endocrinol 1999;50:695-701.

[4] Rossato M, Tavolini IM, Calcagno A, Gardiman M, Dal Moro F, Artibani W. The novel hormone INSL3 is expressed in human testicular Leydig cell tumors: a clinical and immunohistochemical study. Urol Oncol 2011;29:33-37.

[5] Young RH. Sex cord-stromal tumors of the ovary and testis: their similarities and differences with consideration of selected problems. Mod Pathol 2005;18 (Suppl 2):S81-S98.​