Objective To assess the differences between a glucose meter and autoanalyzer at home and broad.Method MEDLINE, CNKI, FMJS, and CBM were searched electronically (1995 to May, 2008). The statistical analysisof included studies was performed according to the Cochrane systematic reviews method. Result Twenty four studies, including 11 English records and 19 Chinese records involving 4 963 specimens, were included in this study. Meta-analysis showed us the blood glucose values of Abbott, Roche, and Johnson abroad subgroups are higher than the laboratory method, and their WMD (95%CI) are 0.57 (0.34,0.80), 0.43 (0.04,0.81), 0.41 (0.11,0.71). The blood glucose values of the Abbot and Roche domestic subgroups are comparable to the laboratory method [WMD= 0.60, 95%CI (– 0.79, 1.99); WMD= – 0.13, 95%CI (– 0.56, 0.29)]. The blood glucose value of the Johnson domestic subgroup is lower than laboratory method [WMD= – 0.95, 95%CI (– 1.42, – 0.48)]. Conclusion The results of the abroad studies are relatively consistent, and the blood glucose values of all abroad subgroups are higher than laboratory method. The domestic studies are different because of other factors.
Objective To assess the quality of diagnostic studies on detecting the tuberculosis antibody to diagnose tuberculosis.Methods CBM (1978 to 2006) and VIP (1994 to 2006) were searched; any author-claimed diagnostic studies which used the dot immunogold filtration assay (DIGFA) to detect the tuberculosis antibody and to diagnose tuberculosis were included. The Quality Assessment of Diagnostic Accuracy Studies (QUADAS) was used to assess the quality of included diagnostic studies by two reviewers independently.Results Thirty-eight papers were included and assessed. We found that most of the quality items were not met with QUADAS. Most papers adopted the retrospective diagnostic case-control design. Thirty-one papers did not describe the selection criteria clearly, 18 did not describe whether all the included patients were verified by using a reference standard of diagnosis, 36 did not describe whether the index test results were interpreted without knowledge of the results of the reference standard, 37 did not report the uninterpretable/intermediate test results, and 34 did not report the withdrawals from the study.Conclusion There are few high quality studies on using DIGFA to detect tuberculosis antibody to diagnose tuberculosis.
A mean of systematic review of diagnostic tests based on The Bayes Library of Diagnostic Studies and Reviews (2nd edition 2002) and Bayes Library are introduced to Chinese readers who are interesting on diagnostic test and screening systematic review.
Objective To evaluate the diagnostic value of ELISA using AgB in cystic echinococcosis (CE). Methods Such databases as PubMed, EMbase, The Cochrane Library, EBSCO, CBM, CNKI, WanFang Data and MedaLink were retrieved on computer, and the relevant journals were also manually searched to collect the trials on ELISA using AgB in diagnosis of CE. The retrieval time was from inception to July 5th, 2012. Two reviewers independently screened the literature, extracted the data and assessed the quality according to QUADAS. Then the meta-analysis was conducted by using Meta-Disc 1.4 software. Results A total of 8 studies were included, and there were 562 CE patients diagnosed by gold standard, 434 suspected cases and 303 healthy people. There were no threshold effects among those 8 studies (the spearman’s correlation coefficient of log sensitivity to log 1-specificity was 0.527, P=0.400 1). The meta-analysis of DerSimonia-Laird showed that, the pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio and diagnostic odds ratio was 0.76 (95%CI 0.73 to 0.79), 0.84 (95%CI 0.82 to 0.86), 5.20 (95%CI 3.59 to 7.55), 0.26 (95%CI 0.18 to 0.35), 23.93 (95%CI 12.35 to 46.39), respectively. And the AUC of SROC was 0.889 7 (Q=0.820 4). Conclusion ELISA using natural AgB and rAgB has greater diagnostic value in detecting CE.
Objective?To evaluate the diagnostic accuracy of procalcitonin (PCT) for ventilator-associated pneumonia (VAP). Methods?We searched MEDLINE, EMbase, The Cochrane Library, CBM, BIOSIS to identify all diagnostic tests which evaluated the diagnostic value of PCT in patients with VAP. QUADAS items were used to evaluate the quality of the included studies. Pooled sensitivity, specificity, positive likelihood ratio (+LR), negative likelihood ratio (-LR), summary receiver operating characteristic (SROC) curve, and the heterogeneity of the included studies were calculated by using the Meta-disk software. Results?Five studies which were identified from 103 references met the inclusion criteria. The summary sensitivity, specificity, +LR, and –LR values were 0.70 (95%CI 0.62 to 0.77), 0.76 (95%CI 0.69 to 0.82), 5.651 (95%CI 1.237 to 25.810), and 0.349 (95%CI 0.155 to 0.784), respectively. Overall area under the curve (AUC) of SROC curve was 0.884 (DOR=19.416, 95%CI 2.473 to 152.47), demonstrating significant heterogeneity (I2gt;50%). Conclusion?The use of PCT for VAP diagnosis has only a moderate sensitivity and specificity. Although the overall accuracy of VAP diagnosis is relatively high, there is significant heterogeneity between the studies, so more high-quality studies are needed. Besides, using PCT alone to diagnose VAP is not sufficient, and a combination with other clinical evaluations is necessary.
With the development of artificial intelligence, machine learning has been widely used in diagnosis of diseases. It is crucial to conduct diagnostic test accuracy studies and evaluate the performance of models reasonably to improve the accuracy of diagnosis. For machine learning-based diagnostic test accuracy studies, this paper introduces the principles of study design in the aspects of target conditions, selection of participants, diagnostic tests, reference standards and ethics.
ObjectiveTo systematically review the diagnostic value of gadolinium ethoxybenzyl diethylenetriamine pentaacetic acid (Gd-EOB-DTPA) for liver metastases.
MethodsWe searched databases including CNKI, CBM, VIP, WanFang Data, PubMed, EMbase and The Cochrane Library from January 2011 to December 2014 to collect diagnostic tests about Gd-EOB-DTPA for liver metastases. Two reviewers independently screened literature, extracted data, and assessed the risk of bias of included studies by using the QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies-2) tool. Then, meta-analysis was performed by using Stata 12.0 software.
ResultsA total of 15 studies from seven countries were included, involving 2 040 nodules from 701 patients. The results of meta-analysis showed that, the pooled sensitivity (Sen), specificity (Spe), positive likelihood ratio (+LR), negative likelihood ratio (-LR) and diagnostic odds ratio (DOR) of Gd-EOB-DTPA for liver metastases were 0.92 (95%CI 0.89 to 0.95), 0.94 (95%CI 0.89 to 0.97), 14.51 (95%CI 8.01 to 26.28), 0.08 (95%CI 0.06 to 0.12), and 177.98 (95%CI 89.50 to 353.94), respectively. The area under curve (AUC) of SROC was 0.97 (95%CI 0.95 to 0.98). The results of subgroup analysis showed that Gd-EOB-DTPA had better Sen in nodules >10 mm than the nodules ≤10 mm in diameter (>10 mm: pooled Sen=0.97, 95%CI 0.94 to 0.99; ≤10 mm: pooled Sen=0.75, 95%CI 0.65 to 0.85; P<0.001); The 3.0T MR had better Sen in diagnosing liver metastases compared with 1.5T MR (3.0T: pooled Sen=0.95, 95%CI 0.92 to 0.97; 1.5T: pooled Sen=0.90, 95%CI 0.87 to 0.94; P<0.001).
ConclusionGdEOB-DTPA is of value for the detection of liver metastases. In particular, it is of high sensitivity for the detection of nodules larger than 10 mm, and for the cases using 3.0T high-field MR system. Due to limited quantity and quality of the included studies, more high-quality studies are required to verify the above conclusion.
Objective To evaluate the diagnostic value of N terminal pro-brain natriuretic peptide (NT-proBNP) for heart failure (HF) and the relationship between NT-proBNP and HF. Methods Applying electrochemiluminescence immunoassay on Elecsys 2 010 quantitatively detects NT-proBNP in HF patients with varing heart damage. Results ① Using 155 pg/ml as the cutoff for diagnosis of HF, the sensitivity and specificity are both 80 %, positive likelihood ratio (PLR) is more than 4.0; when NT-proBNP is more than 350 pg/ml, PLR is more than 14. ② NT-proBNP significantly increases in HF patients，and has significant difference compared with disease control group, Plt;0.05. In dilated cardiomyopathy (DCMP) and acute myocardial infarction (AMI) NT-proBNP is the highest, in coronary heart disease (CHD) and hypertension it’s the smallest. With rising NYHA classes, NT-proBNP increases exponentially. The correlation between NT-proBNP and heart function is good, rs=0.859 (Plt;0.01). Conclusions NT-proBNP for diagnosing HF has a high sensitivity and specificity and can effectively evaluate heart function. With worsening of heart damage, NT-proBNP shows exponential or linear increase.
ObjectiveTo systematically review the value of human epididymis protein 4 (HE4) in early diagnosis of endometrial cancer.
MethodsDatabases including The Cochrane Library (Issue 1, 2013), PubMed, MEDLINE (Ovid), CNKI, CBM and WanFang Data were electronically searched for relevant studies on HE4 versus the golden standard (pathological examination) in the diagnosis of endometrial cancer from inception to April 2013. Meanwhile, relevant journals were also manually searched. Two reviewers independently screened literature according to the inclusion and exclusion criteria, and evaluated the included studies using the QUADAS items. Then, meta-analysis was performed using RevMan 5.1 and Meta-DiSc 1.0.
ResultsFinally, a total of 16 studies involving 2 299 women (1 088 endometrial cancer patients diagnosed according to the golden standard, of which, 504 with benign uterine disease and 707 with normal cervical) were included. The results of meta-analysis showed that, as for HE4 in early diagnosis of endometrial cancer (SEN=57%, 95%CI 0.54 to 0.60; SPE=92%, 95%CI 0.91 to 0.94; +LR=6.92, 95%CI 5.00 to 9.58;-LR=0.46, 95%CI 0.39 to 0.55; DOR=18.38, 95%CI 12.21 to 27.69; AUC=0.881 7).
ConclusionThe current study indicates that serum HE4 is more sensitive and low specific when applied in patients with endometrial cancer, which is worth of being used in clinic. Due to the limitation of low quality of the included studies, more high quality trials are required to verify the above conclusion.
