If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Faias S, Cravo M, Chaves P, et al. A comparative analysis of glucose and carcinoembryonic antigen in diagnosis of pancreatic mucinous cysts: a systematic review and meta-analysis. Gastrointest Endosc. Epub 2021 Apr 11.
The growing rate of pancreatic cyst diagnosis is due to improvements in the quality and increased use of cross-sectional imaging. Better biomarkers that help to discriminate potentially malignant lesions are needed in clinical practice. Previous studies, namely, one study from our group,
showed that cyst glucose concentration performs better than CEA to identify mucinous, potentially malignant cysts, with the need of minimal amounts of cyst fluid. These findings led us to perform this systematic review and meta-analysis.
and as such it should have not been considered. We included it, but we registered the information of the surgical cohort of this study as NA (nonavailable) in Table 2 of our article, and in the quality assessment using QUADAS-2 (Fig. 2A of our article), we classified it with unclear risk of bias for patient selection.
Faias S, Cravo M, Chaves P, et al. A comparative analysis of glucose and carcinoembryonic antigen in diagnosis of pancreatic mucinous cysts: a systematic review and meta-analysis. Gastrointest Endosc. Epub 2021 Apr 11.
We correctly refer the 275 patients with surgical pathology specimens available, and only those should have been included in the final analysis. Because of this, we have redone the meta-analysis, including the forest plots of sensitivity and specificity (Fig. 1), the receiver operating characteristics (ROC) curve (Fig. 2), the sensitivity analysis (Fig. 3), and Deeks’ funnel plot (Fig. 4) for glucose studies. The results, including only 4 studies with surgical pathologic diagnosis, confirm all previous findings with 5 studies, with identical sensitivity of 0.90 (95% confidence interval [CI], 0.85-0.94) and improved specificity of 0.82 (95% CI, 0.72-0.89) and area under the ROC curve of 0.96.
Figure 1Forest plots of the studies for glucose. Between brackets, the 95% confidence intervals (CI) of the sensitivity and specificity. The figure shows the estimated sensitivity and specificity of the study (red circle) and its 95% CI (blue horizontal line). The area of the circle reflects the weight that the study contributes to the meta-analysis.
that using surgical pathology as a reference standard for diagnosis limits the data available to evaluate the sensitivity and specificity of glucose, but we chose to do so because histologic analysis is the criterion standard for diagnosis.
Although 3 studies in the glucose group were co-authored by Dr Park from Stanford University, in the study by Carr et al,
the samples were obtained at Indiana University Health University Hospital, and there was no overlap of patients with those in the studies by Park et al
In those 2 studies, samples were obtained from the bio-repository for pancreatic cyst fluid at Stanford Medical Center, and there is an overlap of 19 patients in a total of 110 patients included. The limited number of publications, and the low number of patient overlap, led us to accept both studies.
Faias S, Cravo M, Chaves P, et al. A comparative analysis of glucose and carcinoembryonic antigen in diagnosis of pancreatic mucinous cysts: a systematic review and meta-analysis. Gastrointest Endosc. Epub 2021 Apr 11.
the data of CEA and glucose levels of each study were extracted to Table 2, and a cutoff level of CEA 192 mg/mL was used for calculations. In the 9 studies included using a different CEA cutoff level, it was possible to extract individual CEA levels from each study. As an example, in the study by Frossard et al
Faias S, Cravo M, Chaves P, et al. A comparative analysis of glucose and carcinoembryonic antigen in diagnosis of pancreatic mucinous cysts: a systematic review and meta-analysis. Gastrointest Endosc. Epub 2021 Apr 11.
a total of 8 full-text articles as presented in Figure 1 of the article, because individual CEA levels were not presented.
Our aim was to evaluate the utility of pancreatic cystic fluid glucose for mucinous cyst diagnosis in EUS-FNA samples before surgery, not because glucose is a biomarker to select patients for surgery, but because the surgical pathologic specimen is the criterion standard for cyst diagnosis. Glucose and CEA are both biomarkers of mucinous cysts, with additional cytologic or morphologic features required to select cysts for surgery.
glucose level was evaluated with a glucometer in 7 cystic lesions, both in a fresh sample in the EUS room and later in a frozen sample, as presented in Table 3 in the article. Nonsignificant changes of glucose level occurred, never crossing the 50 mg/dL cutoff level.
We believe that although this case is not settled, our meta-analysis clearly demonstrates the advantage of using glucose levels measured in the room by use of a glucometer, over CEA for mucinous cyst diagnosis.
Disclosure
All authors disclosed no financial relationships.
References
Confer B.D.
Iqbal U.
Diehl D.L.
Comparison of pancreatic cystic fluid glucose and carcinoembryonic antigen in the diagnosis of pancreatic mucinous cysts.
Faias S, Cravo M, Chaves P, et al. A comparative analysis of glucose and carcinoembryonic antigen in diagnosis of pancreatic mucinous cysts: a systematic review and meta-analysis. Gastrointest Endosc. Epub 2021 Apr 11.
We read with interest “A comparative analysis of glucose and carcinoembryonic antigen in diagnosis of pancreatic mucinous cysts: a systematic review and meta-analysis” by Faias et al.1 The authors conclude that pancreatic cyst fluid (PCF) glucose is more accurate than carcinoembryonic antigen (CEA) in determining whether the lesion is mucinous. There are several limitations in the reviewed body of literature that make this conclusion suspect. One of the 5 studies did not compare glucose with surgical pathologic analysis and should not have been included in the meta-analysis, given the inclusion criteria set by the authors.
01171-8&id=gr1.jpg){kind=link}
01171-8&id=gr2.jpg){kind=link}
01171-8&id=gr3.jpg){kind=link}
01171-8&id=gr4.jpg){kind=link}