(ie ischemia) in patients suspected for CGI. The technique is

Endoscopic visible light spectroscopy: a new minimally invasive technique for the
diagnosis of chronic gastrointestinal ischemia
D. van Noord MD1, D.A. Benaron MD2, P.M.T. Pattynama MD PhD3, H.J.M. Verhagen MD
PhD4, E.J. Kuipers MD PhD1, P.B.F. Mensink MD PhD1
Departments of Gastroenterology and Hepatology1, Intervention Radiology3, Vascular
Surgery4, Erasmus MC - University Medical Center, Rotterdam, The Netherlands
Standford University School of Medicine, Palo Alto, CA, USA2
The diagnosis of chronic gastrointestinal ischemia (CGI) remains challenging. Currently,
there is no single and simple test with a high sensitivity available for detection of CGI. Visible
light spectroscopy (VLS) is a new technique that non-invasively measures actual mucosal
capillary hemoglobin oxygen saturations during endoscopy (T-Stat 303 Microvascular Tissue
Oximeter, Spectros Corporation, California). This saturation reflects the adequacy of mucosal
perfusion and is therefore lowered in CGI. A recent pilot study using VLS in a few CGI
patients showed very promising results in detection of mucosal hypoxemia. In this study we
evaluated the actual diagnostic value of VLS for detection of ischemia in a large cohort of
patients clinically suspected of CGI.
Prospectively, consecutive patients referred for evaluation of possible CGI were included.
Patients underwent endoscopy-guided VLS next to the standard work-up consisting of a
combination of gastrointestinal tonometry (TM) and CT or MR abdominal angiography. VLS
measurements were performed at 5 standard locations (distal oesophagus, corpus, antrum,
bulbus and duodenum) during upper endoscopy. VLS was performed before start of TM and
radiological diagnostics. Mucosal saturation measurements during VLS were compared with
the diagnosis CGI and results of TM.
In 11 months, 80 patients were included: 30 males, mean age 59 (17 – 86) years. CGI was
diagnosed in 58 (73%) patients: 26 single-, 20 multi-vessel disease, and 12 patients with nonocclusive CGI. The VLS mucosal saturations during endoscopy in means ± SD were for CGI
and non-CGI patients respectively: 60.2 ± 4.0 and 62.3 ± 5.3 (oesophagus), 62.1 ± 5.6 and
64.9 ± 3.4 (corpus), 63.0 ± 5.4 and 65.6 ± 3.8 (antrum), 58.9 ± 5.2 and 62.6 ± 4.4 (bulbus),
54.4 ± 6.7 and 59.2 ± 4.1 (duodenum), 59.6 ± 3.4 and 62.8 ± 2.3 (overall). The mean
saturation in the antrum, bulbus, duodenum and the overall mean saturation were significantly
lower in CGI patients as compared to non-CGI patients. Accordingly, low saturation levels
measured with VLS correlated significantly with abnormal TM results (p<0.0001). No
differences were found comparing saturations in single- or multivessel disease CGI patients.
Conclusions: VLS measurement of mucosal oxygenation during gastroduodenoscopy is a very
promising technique for detection of actual mucosal hypoxemia (i.e. ischemia) in patients
suspected for CGI. The technique is easy to perform and shows excellent correlation with