{"paper_id":"0e790222-4191-4bf5-8e8f-35edcad4e9e5","body_text":"R E S E A R C H Open Access\nEndogenous fluorescence of hemosiderin\nin endometriosis to improve clinical\ndetection\nAndrew G. Cabe 1,6* , Arnold D. Estrada 3, Taylor Hoyt 1, Xiao Yang 1, Scott Jenney 3, Philip T. Valente 4, Bryan Cox 5,\nJessica E. McLaughlin 2, Randal D. Robinson 2, Thomas E. Milner 3 and Marc D. Feldman 1\nAbstract\nBackground: Endometriosis impacts 6–10% of all reproductive- age women. Studies have shown the more effectively\nendometriosis is removed, the better the patient outcomes for pain reduction and fertility (2, 3). Hemosiderin, glands,\nand stroma are the histologic markers of endometriosis; optical coherence tomography (OCT) can identify glands and\nhemosiderin has a known endogenous fluorescence than can be detected by two-photon microscopy (TPM). The\nhypothesis was that the identification of optical properties of endometriosis using OCT and TPM combined would\nimprove a surgeon ’s ability to diagnose and treat by improving endometriosis detection compared to current\nstandards of visual diagnosis.\nMethods: Forty-one women with clinically suspected endometriosis undergoing laparoscopy were consented. Women\nwere enrolled at two clinical sites: University of Texas He alth Science Center, San Antonio and Methodist Healthcare\nSystem, San Antonio. The surgeon made a clinical diagnos is of suspected endometriosis as 1) yes present 2) maybe\npresent, and 3) not present (controls) from the peritoneum without suspected disease. One-hundred-twenty biopsies\nwere collected from 27 women with visual ly suspected endometriosis. All three patient biopsy classes were excised and\nunderwent histologic examination as the gold-standard diagnosis for endometriosis. The samples were imaged ex-vivo\nfor optical markers of endometriosis; OCT for endometrial glands and TPM for hemosiderin. Histologic markers were co-\nregistered with optical properties. Biopsies were embedded in agar to maint ain orientation du ring imaging and\nhistological processing. TPM used the endogenous fluorescence of hemosiderin as a marker. OCT used glands as a\nmarker. Sensitivity, specificity, and positive (PPV) and negative predictive values (NPV) were calculated.\nResults: The main-outcome-measure was the statistical comparison of clinical impression, imaging results, and histologic\ntruth. Glands, stroma and hemosiderin were present in 49, 72 and 86% of endometriosis samples confirmed by histology.\nClinical suspicion of endometriosis had 98% sensitivity, 53% specificity, 68% PPV, and 96% NPV. In 31 samples of\nendometriosis maybe being present, 39% were histologically confirmed. Eighty-eight samples were analyzed using OCT-\nTPM. OCT-TPM had 93% sensitivity, 100% specificity, 100% PPV, and 93% NPV.\nConclusions:OCT-TPM is useful in identifying endometriosis’ presence or absence. Evaluation of suspected endometriosis\nby OCT-TPM improves surgeons’ abilities to diagnose and treat endometriosis.\nKeywords: Endometriosis, Optical coherencetomography, Auto-fluorescence\n© The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0\nInternational License ( http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and\nreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to\nthe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver\n(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.\n* Correspondence: cabe@uthscsa.edu\n1Departments of Cardiology, University of Texas Health Science Center at San\nAntonio, 7703 Floyd Curl Drive, San Antonio, TX 78229, USA\n6Department of Medicine, University of Texas Health Science Center, 7703\nFloyd Curl Drive, San Antonio, TX 78229, USA\nFull list of author information is available at the end of the article\nTranslational Medicine\nCommunications\nCabe et al. Translational Medicine Communications             (2019) 4:9 \nhttps://doi.org/10.1186/s41231-019-0038-3\n\nBackground\nEndometriosis impacts 6 –10% of all reproductive- age\nwomen [1]. The gold standard for diagnosis and treatment\nof endometriosis is laparoscopic inspection and excision or\nablation with histologic confirmation. Studies have shown\nthe more effectively endometriosis is removed, the better\nthe patient outcomes for pain reduction and fertility [ 2, 3].\nHowever, of the tissue removed during laparoscopy, less\nthan 70% is confirmed as endometriosis on histologic diag-\nnosis [4]. With a 30 –40% recurrence rate for symptomatic\nendometriosis after laparoscopic surgery, due to incomplete\nsurgical excision [5], there are clearly short-comings in the\ncurrent surgical diagnosis and management of endometri-\nosis. During laparoscopic surgery, the surgeon’sg o a li si d e n -\ntification of each endometriosis lesion and sequential\nexcision of each implant. Complicating the time consuming\nprocess of identification and removal is the potential wide-\nspread distribution of endometriosis lesions throughout the\nabdominal and pelvic cavities. The distribution of these le-\nsions presents a challenge to all surgeons whose diagnostic\nimaging tools are normally limited to an unaided eye and\nlaparoscope with some endometriosis not being immedi-\nately discernible from surrounding tissue. The nature of the\niceberg geometry of endometriosis lesions complicates the\nsurgery since frequently only a fraction of the lesion is vis-\nible with the remaining endometriosis lesion located below\nthe surface. Because the surgeon is faced with the divergent\nchallenges of removing all endometrial lesions without dam-\naging delicate underlying tissues and structures, prolonged\nsurgeries are common and only partially successful. Patient\noutcomes suffer because no effective imaging tools are cur-\nrently available in the operating room to allow endometri-\nosis lesions to be discerned from surrounding tissues [6].\nThe current study presents novel solutions for the de-\ntection of endometriosis, using optical coherence tomog-\nraphy and two photon microscopy (OCT-TPM) imaging.\nHemosiderin is known to have specific excitation/emis-\nsion wavelengths [ 7, 8]. We hypothesized that we could\nidentify hemosiderin as a marker in endometriosis with\nTPM imaging of the same excitation/emission pair. He-\nmosiderin is particularly relevant because is present in\nearly-onset endometriosis [ 9]. Further, it is known that\ncolumnar epithelial cells, found in endometriosis glands,\nform fluid filled glandular structures and OCT is an op-\ntical technique capable of easily identifying fluid filled\nspaces in tissues [ 10]. Therefore, we hypothesized that\nOCT could be a second optical tool in the detection of\nendometriosis. We tested these hypotheses on ex vivo\ntissue samples collected by gynecologic surgeons from\nwomen suspected of having endometriosis. Samples\nwere classified by the surgeon as yes or maybe for the\npresence of endometriosis, with suspected negative con-\ntrols also excised, and these samples were then evaluated\nwith both optical techniques and compared to the gold\nstandard, histologic diagnosis. The histologic diagnosis\nwas made by a gynecologic pathologist (PTV) blinded to\nthe results of the OCT-TPM imaging.\nMethods\nThe tissue samples were collected at two clinical sites,\nUniversity of Texas Health Science Center at San Anto-\nnio, and Methodist Healthcare System in San Antonio.\nParticipants were enrolled between December 18, 2014\nand June 6, 2017 in a prospective manner. Two experi-\nenced gynecologic surgeons (RDR, BC), each with\ngreater than 25 years ’ experience diagnosing and treating\nendometriosis performed all of the surgical procedures.\nEthical approval\nIRB approval for this study was obtained at both clinical\nsites. Background information was collected on patients ’\nage and race, medical history, and list of medications.\nSubjects were approached for consent if they were\nundergoing surgery for suspected endometriosis for clin-\nical indications. Informed consent was obtained for all\nparticipants. Subjects were withdrawn if no suspected\nendometriosis was identified at the time of surgery. The\nmean participant age was 37 (range 22 –46).\nSample collection\nPeritoneal biopsy samples were collected at the time of\nlaparoscopic surgery and the surgeons were asked to make\na clinical diagnosis of each sample removed as “yes”, “no”,\nor “maybe” for endometriosis. The location of origin of\nthese samples was also recorded. Clinically defined endo-\nmetriosis was described as red, black, or white lesions.\nThe samples designated in the “no” category were control\nsamples of normal appearing peritoneal tissue biopsied by\nthe surgeon. These three categories allowed us to develop\noptical signatures of endometriosis. The “maybe” category\nallowed us to determine if these optical properties could\nimprove the sensitivity and specificity of the surgeon ’s\nclinical diagnosis. The “maybe” samples represent sus-\npected disease that some surgeons might remove and\nothers might not depending on location and appearance.\nOnce the samples were received, imaging was per-\nformed to categorize optical properties. Samples were\nembedded in agar using a 4x4x2 mm mold and imaged\nin the mold. The mold allowed for co-registration with\nhistology. Specifically ,p r e - e m b e d d i n gi nam o l d\nallowed the preservation of tissue orientation and lo-\ncation throughout imaging an d histological processing.\nKnowledge of tissue orientation allows exact 3D\nco-registration between the OCT-TPM images and\nstained histologic slides.\nCabe et al. Translational Medicine Communications             (2019) 4:9 Page 2 of 11\n\nImaging procedures\nOptical imaging was performed with three imaging sys-\ntems, Optical Coherence Tomography (OCT), Two-Pho-\nton Microscopy (TPM), and Wide-Field/Single-Photon\nFluorescence. Imaging was performed within 24 h of sam-\nple excision. OCT imaging was performed on a benchtop\nOCT system built in our lab, using Alazar (Pointe-Claire,\nQC, Canada) and Santec (Hackensack, NJ, USA) sources,\nwhich operate between 1250 and 1376 nm. These systems\nrecorded b-scans for the entire sample volume (up to\n6.25 × 6.25 × 2 mm). Lateral resolution was 6.1 –12.21 μm/\npixel. Axial resolution was 3.75 –8 μm/pixel. The targets\nfor OCT imaging were glands. The presence of endomet-\nriosis detected by OCT was a void in the image which cor-\nresponded to a fluid-filled gland in the histology. OCT\npenetration depth and wide-field also provided a full vol-\nume view of the samples, ranging in size from 12 to 27\nmm3, useful for later co-registration with histology. The\nsecond optical target for endometriosis was the intrinsic\nfluorescence of hemosiderin. Hemosiderin has an excita-\ntion peak of 450 nm, and an emission peak near 680-690\nnm. TPM imaging and Single Photon fluorescence im-\naging were performed on a Prairie View microscope sys-\ntem (Middleton, WI, USA). TPM was performed using an\nexcitation wavelength of 900 nm (twice the one-photon\nexcitation peak). Fluorescence images using emission\nchannels, 500 nm, 560 nm, 620 nm, and 690 nm, were re-\ncorded. Hemosiderin was identified by increased pixel in-\ntensity in the 690 nm emission channel. TPM images were\ntaken with a field of view of 735 μm. Z-series with 10 μm\nstep-size were taken and stitched together to image the\nentire sample. Z-series (stacks of images) were taken in\ndepth until the image was no longer interpretable (ap-\nproximately 200 μm). TPM had a resolution of 1.44 μm/\npixel. Additionally, single-photon wide-field fluorescence\nimages were recorded using an excitation source emitting\nat 420 + − 20 nm and an emission filter of 690 + − 20 nm\nto demonstrate hemosiderin detection with standard\nfluorescence imaging as seen in increased pixel intensity.\nThis method is less expensive than TPM and therefore\nmore translatable to a clinical device.\nHistologic processing\nThe slides underwent histologic processing. Slides were\ntaken every 40 μm for the first 400 μma n de v e r y1 0 0μm\nsubsequently. Paraffin slides were stained with Hematoxylin\nand Eosin. For the histologic evaluation, each sample was\ncategorized for the presence of the three histologic markers\nfor endometriosis: glands (columnar endothelial cells),\nstroma, and hemosiderin. The histologic diagnosis was used\nas the gold standard aga inst which the surgeons ’ and im-\naging systems’ findings were compared. Due to the exact-\nness of the match from the use of agar, the optical\nproperties of specific markers were obtained. Unique\nfeatures in the images were identified as matching the\nhistologic markers.\nStatistical analyses\nSensitivity, specificity and negative and positive predict-\nive values for the accuracy of both the imaging systems\nand the surgeons ’ impressions were calculated. Sensitiv-\nity was calculated by taking the number of true positives\n(histologically confirmed endometriosis) and dividing by\nthe sum of the number of true positives and false nega-\ntives (histologically confirmed endometriosis that was\nnot detected by the surgeon or imaging systems). Speci-\nficity was calculated by taking the number of true nega-\ntives (histology confirmed to lack endometriosis) and\ndividing by the sum of the number of true negatives and\nfalse positives (histology confirmed to lack endometriosis\nbut identified as disease positive by the surgeon or im-\naging system). Positive predictive value was calculated by\ntaking the number of true positives and dividing by the\nsum of true positives and false positives. Negative pre-\ndictive value was calculated by taking the number of true\nnegatives and dividing by the sum of true negatives and\nfalse negatives.\nResults\nOCT was used to identify fluid filled glands and two-pho-\nton microscopy (TPM) to detect the endogenous fluores-\ncence of tissue without exogenous labels, specifically\nhemosiderin. Results show that glands are visible via OCT\nimaging (Fig. 1a) and hemosiderin has endogenous fluor-\nescence that is present both in multi-photon (Fig. 1b) and\nsingle-photon fluorescence imaging (Fig. 1c).\nClinical impression\nThis study enrolled 41 women. Based on the clinical\ndiagnosis of the physician, 27 women had endometriosis\nand 14 did not. These 14 women were determined to\nhave no endometriosis lesions by the surgeon and no bi-\nopsies were obtained for ethical reasons. The results out-\nlined below focus on the remaining 27 subjects. Of these\n27 subjects, 18 had advanced optical imaging performed\nwith OCT and TPM.\nHistology\nFor the set of 27 subjects the surgeon ’s diagnosis of\n120 biopsies was compared to the histologic gold\nstandard. The surgeon diagnosed 65 of these as defin-\nite endometriosis, 31 as maybe for endometriosis, and\n24 as negative for endometriosis. Fifty-seven biopsies\nwere interpreted blindly by the gynecologic patholo-\ngist to have endometriosis; 44 of the definite biopsies,\n12 of the maybe biopsies and one of the negative bi-\nopsies. The accuracy of the surgeons ’ clinical impres-\nsion is summarized in Table 1.\nCabe et al. Translational Medicine Communications             (2019) 4:9 Page 3 of 11\n\nHemosiderin was a prevalent marker seen in 49 of 57\n(86%) endometriosis samples proven positive by histology.\nGlands were found in pathology in 28 of the 57 (49%) of\nendometriosis positive samples. The final marker, stroma,\nwas found in 41 of the 57 samples (72%), although not an\noptical target for advanced imaging.\nAdvanced optical imaging\nIt was possible to obtain full optical imaging sets for 18 of\nthe 27 patients outlined above. This resulted in 113 optical\nimaging sets of co-registered OCT, TPM and histology from\n86 biopsy samples. Results are shown in Tables1, 2,a n d3.\nTPM confirmed the presence of hemosiderin with a\nvisible endogenous fluorescence signal in all 36 of those\nsamples where it was confirmed by histology and did\nnot have a single false positive in the other 50 samples\n(Fig. 2). Glands were visible in OCT for 17 of the 21\nsamples for which they were seen in histology. OCT for\nglands lacks a PPV and specificity due to lack of a true\nfalse positive. This is due to OCT ’s optical marker for\nglands being able to be co-registered with its histologic\ncounterpart, but it is difficult to examine the OCT image\nand read a false positive that corresponds to histology.\nSingle photon results\nTo allow for clinical translation of hemosiderin as an op-\ntical target, 4 subjects were further examined by an add-\nitional imaging technique, single-photon fluorescence.\nFig. 1 Imaging Results. 1 a Sample 29-1a, the top image is the OCT en face image, the bottom image is the histology, the identified gland is\nboxed in red, scale bar is 1 mm 1 b Sample 20–6, TPM image on top, histology on bottom, hemosiderin is circled in red and appears as brown in\nthe histology due to H&E stain, the TPM image is colored based on emission channel, the red represents 690 nm (hemosideran), blue is 500 nm\n(collagen), scale bar is 1 mm 1 c: Sample 34 –1 The image on the top left is the single photon microscopy/ wide field fluorescence, very bright\nspots match with TPM and histology as hemosiderin, the image on the top right is the corresponding TPM image location colored as stated in\n1b these are scale to 470 um × 470 um, the bottom right is the full TPM image and bottom left is the corresponding histology, the red box\nrepresents the area of the top images, scale bar is 1 mm\nTable 1 Statistical results of surgical impression/ advance optical imaging vs histologic gold standard\nPPV NPV SENS. SPEC.\nSurgical Impression (27 subjects) 68 96 98 53\nOCT-TPM Combined (18 subjects with advanced imaging) 100 93 93 100\nTPM for Hemosiderin (113 optical imaging sets) 100 100 100 100\nOCT for Glands (113 optical imaging sets) 96 82\nPPV is positive predictive value, NPV is negative predictive value, SENS is sensitivity, SPEC is specificity\nSurgical Impression is based on the surgical diagnosis of all samples taken from 27 subjects against histological diagnosis. OCT-TPM combined is the optical\nmarkers from both imaging type predication of endometriosis against histological diagnosis. OCT-TPM was only available in 18 of 27 patients with endometriosis.\nOne hundred thirteen optical imaging sets were available from these 18 patients for OCT-TPM. OCT for glands lacks a PPV and specificity due to lack of a true\nfalse positive. TPM for hemosiderin and OCT for glands is specifically the identification of the marker in the imaging data against histologic confirmation of\nthe marker\nCabe et al. Translational Medicine Communications             (2019) 4:9 Page 4 of 11\n\nThis evaluation examined 6 biopsies from these 4 pa-\ntients identified by TPM to be hemosiderin positive, and\n2 controls shown by TPM to be hemosiderin negative.\nSingle-photon was able to detect hemosiderin similar to\nTPM, in the exact areas and identical pattern as TPM,\nand was confirmed by histology (Fig. 1c).\nDiscussion\nOur results demonstrate the benefit of adding advanced\noptical imaging in the diagnosis of endometriosis. The\nclinician made a correct clinical diagnosis in 67% of\nhistologic positive samples. Endometrial glands are help-\nful if present but were only found in histology in 28 of\nthe 57 endometriosis positive samples or 49% of the\ntime. Glands were visible in OCT for 19 of the 23 im-\naging sets for which they were seen in histology. The sig-\nnificant finding was the frequency that hemosiderin was\npresent. Hemosiderin was found in 86% percent of all\nhistologic samples, and correctly identified with TPM\n100% of the time; as well as, in single photon fluores-\ncence imaging, which will allow clinical translation of\nhemosiderin identification to the operating room.\nSpecifically, these data demonstrate the need for a bet-\nter detection method than a surgeon ’s clinical diagnosis.\nOf the 65 samples of endometriosis removed where the\ngynecologic surgeon was certain that endometriosis was\npresent, the surgeon was incorrect 21 times, confirming\nprior reports of the relatively poor visual diagnosis of\nendometriosis lesions during laparoscopic surgery for\nendometriosis. We had two experienced gynecologic sur-\ngeons at two institutions and generated 120 total tissue\nsamples for evaluation. Our data is comparable with\nresults from studies with larger cohorts, Mettler re-\nported a 53.8% positive rate in biopsies of suspected\nendometriosis [ 11] and Stratton a 61% rate [ 12]. Steg-\nmann reported in her 133 patient, 611 sample study, a\ntrue positive rate of 65.0% rate which is very similar to\nour 67% rate but she found a much higher false negative\nrate of 12% compared to our 4% [ 4].\nThere are several implications of our results regarding\nsurgeons removing too much tissue in 1/3 of cases and\nmissing lesions that should be removed. If any endometri-\nosis is being missed and left in the patient, the chances for\nimproving patient outcomes of pain reduction and pos-\nsibly increased fertility, and preventing the high persist-\nence rates common in endometriosis, are diminished. The\nlow specificity and PPV reveals how much non-endomet-\nriosis tissue the surgeons are removing during the proced-\nure and how much unproductive surgical time is added to\nthe procedure. The one false negative result reveals that\nthere is diseased tissue that is being left behind because\nthe physician could not determine that a negative biopsy\nwas actually endometriosis. This study also examined\n“maybe” samples, where the surgical diagnosis was un-\nclear. These samples proved to be positive for endometri-\nosis 39% of the time. Our combined TPM-OCT imaging\naccurately classified every “maybe” case. This shows the\nadditive value of an advanced optical imaging approach,\nwhich would assist the surgeon in accurately identifying\nendometriosis where it is unclear whether endometriosis\nis present or not.\nTPM-OCT detection overcomes many of the limita-\ntions of other methods developed to improve detection.\nStudies have shown that specific endometriosis lesions\nTable 2 Clinical Impression vs. Histology Gold Standard\nPhysician\nclinical\nimpression\nat time of\nsurgery\nSubjects\n7 8 9 1 01 11 21 31 5 1 6 2 02 22 3 2 4 2 52 72 82 93 03 23 43 63 83 94 24 54 64 7\nYes No No No Yes Yes No No Yes Yes No No Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes Yes\nNo No Yes Yes Yes No No Yes Yes No Yes Yes Yes Yes No Yes Yes No Yes Yes\nYes Yes No Yes Yes No Yes No No Yes\nYes Yes No Yes No No\nYes Yes\nYes Yes\nMaybe No Yes No Yes No Yes No No No No No No Yes Yes Yes Yes No Yes Yes Yes\nNo No Yes Yes No No No No No\nNo\nNo\nNo No No No Yes No No No No No No No No No No No No No No No No No No No No\nHistological diagnosis\nBlank = no sample obtained; each yes or no entry represents each individual sample of tissue biopsied; columns represent the histological diagnosis per the\nreading of an expert Ob/Gyn pathologist; rows represent Ob/Gyn physician/surgeon clinical impression of the sample at the time of surgery\nCabe et al. Translational Medicine Communications             (2019) 4:9 Page 5 of 11\n\nTable 3 Full Imaging Results\nSubject Sample Subsec. CI Type Path. TPM-hemo OCT-glands\n20 1 Yes Red H Yes\n2 Yes Red H Yes\n3 b Yes Red H,S Yes\n4 b Yes Dark chocolate cyst H Yes\nc Yes S No\n5 a Yes Cyst H Yes\nc Yes H,S Yes\nd Yes H Yes\n6 Yes Endometroma H Yes\n22 1 a Yes Red+black – No\nb Yes G,S No No\n2 No Control – No\n23 1 No Control – No\n2 a Yes Red – No\n3 a Yes red – No\nb Yes – No\n4 b Yes Black+white – No\n24 1 a Yes Red – No\nb Yes – No\n2 Yes Black + white – No\n3 No Control – No\n25 1 a Yes Red S No\nb Yes G,S No No\n2 a Yes Black H,S Yes\nb Yes H,S Yes\n3 No Control – No\n27 1 a Yes Red + black G,S No Yes\nb Yes – No\n2 Maybe White – No\n3 No Control – No\n4 a Yes Endometroma H Yes\nb Yes H,S Yes\nc Yes H Yes\n5 Yes Black H Yes\n6 Yes Red – No\n28 1 Maybe White – No\n2 a Yes Red H Yes\nb Yes – No\nc Yes H Yes\n3 a Yes Red – No\nb Yes – No\nc Yes – No\n4 No Control – No\n29 1 a Yes Red G,H,S Yes Yes\nCabe et al. Translational Medicine Communications             (2019) 4:9 Page 6 of 11\n\nTable 3 Full Imaging Results (Continued)\nSubject Sample Subsec. CI Type Path. TPM-hemo OCT-glands\nb Yes G,H,S Yes Yes\n2 No Control – No\n3 a Yes Red + black – No\nb Yes G,H,S Yes Yes\n4 Maybe Black – No\n30 1 Yes Black H Yes\n2 maybe White – No\n3 a Yes Black H Yes\nb Yes H Yes\n4 Yes E, H Yes\n5 a Yes H Yes\nb Yes E,H Yes\n7 No Control – No\n32 1 Yes Red G,H,S Yes Yes\n2 Yes Red G,H,S Yes Yes\n3 Maybe Black E,H,S Yes\n4 Maybe White G,H,S Yes Yes\n5 No Control – No\n34 1 Yes Black G,H,S Yes Yes\n2 a Yes Red G,H,S Yes Yes\nb Yes G,S No Yes\n3 Maybe White H, S Yes\n4 No Control – No\n36 1 No Control – No\n2 a Maybe Black H,S Yes\nb Maybe – No\n3 a Yes Red G,S No No\nb Yes E,H,S Yes\n4 Maybe Black + white H,S Yes\n38 1 Yes Red G,S No Yes\n2 Yes Red – No\n3 Yes Black – No\n4 Maybe White H Yes\n5 No Control – No\n6 Maybe White – No\n39 1 Yes S No\n2 Maybe – No\n3 No Control – No\n4 Maybe – No\n5 a Yes H,S Yes\nb Yes H Yes\nc Yes – No\nd Yes – No\n42 1 No – No\nCabe et al. Translational Medicine Communications             (2019) 4:9 Page 7 of 11\n\nare not easily identified laparoscopically or by CT, MRI\nor ultrasound, unless there are large masses [ 13, 14].\nThis limits the surgical relevance of these technologies.\nLimited success in detecting endometriosis has been\nseen in comparing blue light adsorption of surrounding\ntissue with endometriosis, which does not absorb blue\nlight. However blue light techniques do not provide a\nmechanistic link with a histological gold standard, such\nas stroma, glands, or hemosiderin and thus its value to\nclinicians has been questioned and not adopted [ 15].\nOther imaging methods use exogenous materials such as\nthe photo-sensitizer 5-ALA, Firefly ™ fluorescence im-\naging (ICG), and blue dye [ 15, 16]. All have had limited\nsuccess in identifying markers associated with endomet-\nriosis, including angiogenesis and enhanced vascularity,\nchanges in the peritoneal tissue surrounding endometri-\nosis, and continued epithelial bleeding. However, the in-\ndirect nature of these approaches has also limited\nwidespread adoption. Thus, newer techniques are\nneeded that are both diagnostic based on the true histo-\nlogic pathology of endometriosis, and marker free as will\nbe shown in a more detailed discussion of these compet-\ning optical techniques below.\nDemco and co-workers published [ 17] an endometri-\nosis blue light reflectance/absorption imaging method.\nThe authors hypothesized that lesions not visible under\nwhite light were visible under blue light reflectance be-\ncause porphyrin molecules in endometriosis absorb blue\nlight. This is not an auto fluorescence technique as it\nlooks at absorption of blue light and not emission from\nexcitation. Our TPM detection of hemosiderin looks at\nthe emission signal from excited hemosiderin. Demco ’s\nmethod has not been rigorously tested since there are\nno sensitivity and specificity results published [ 17]. Fur-\nthermore, the spectral analysis of Demco ’s technique will\nnot reveal endometriosis below the surface such as\nTable 3 Full Imaging Results (Continued)\nSubject Sample Subsec. CI Type Path. TPM-hemo OCT-glands\n2 Maybe – No\n3 Yes – No\n4 a Maybe G, H, S Yes Yes\nb Maybe – No\n5 Yes G, S No Yes\n6 Yes G, H, S Yes Yes\n7 Yes G, S No Yes\n45 1 Maybe – No\n2 Maybe H, S Yes\n3 Yes G, S No Yes\n4 Yes – No\n5 Yes – No\n6N o – No\n46 1 Yes H Yes\n2 a Yes – No\nb Yes – No\n3 a Yes G, H, S Yes Yes\n4N o – No\n47 1 Maybe G, S No Yes\n2 Yes G, H, S Yes No\n3 Yes G, H, S Yes Yes\n5 Yes – No\n6 Maybe – No\n7 Yes H, S Yes\n8N o – No\nColumns: Subject: Subject number, Sample: sample number from that subject, Subsec.: is the subsection if the sample had to be divided to be fully imaged due\nto sample size, CI (Clinical Impression): yes is positive for endo, maybe is possible endometriosis normally at the surgeon ’s discretion for removal, no is control\ntissue, Type: the surgeon reading of type of endometriosis, Path: Pathology reading - G is gland, fluid filled sac, E is endothelial cells with no fluid filled sac, H is\nhemosiderin, and S is stroma. TPM-hemo: yes is a present optical marker for hemosiderin, no is no optical marker as determined from the TPM image, OCT-glands:\nyes if the glands is seen and co-registered between the OCT image and histology, no if no such co-registration was possible\nCabe et al. Translational Medicine Communications             (2019) 4:9 Page 8 of 11\n\nunrecognized intra-ovarian endometriosis, since it relies\non surface reflection. Our technique depends on en-\ndogenous fluorescence emission and will identify lesions\nbelow the surface since we are detecting emitted pho-\ntons which can pass through layers of tissue.\nBuchweitz et al. developed an auto-fluorescence tech-\nnique to detect differences between normal tissue and\nendometriosis [ 18]. They hypothesized that well known\nauto-fluorescence of NADH, and ATP has a different\nfluorescence pattern in endometriosis tissue compared to\nnormal tissue. The reasoning behind their approach is that\nendometriosis has a different metabolic activity than sur-\nrounding tissues, and different metabolic rates can be re-\nvealed in auto-fluorescence imaging due to the different\nendogenous fluorescence ’s of NAD+ and NADH, and\nATP and ADP. With Buchweitz ’s method endometriosis\nwas identified by decreased fluorescence compared to sur-\nrounding tissue, looking for a reduction in auto- fluores-\ncence as the marker for endometriosis. The emission\nwavelengths of those molecules are shorter than the\nemission wavelength of the porphyrin in hemosiderin,\nwhich might explain why they did not see the enhanced\nfluorescence from hemosiderin as we did [ 7, 8]. Looking\nfor a reduction in fluorescence is flawed in that this ap-\nproach detects a negative image or the absence of a signal.\nAnother approach utilizes exogenous fluorescence\nwith application of δ-Aminolevulinic acid (5-ALA)\nwhich was originally developed for photodynamic ther-\napy in cancer. It has been applied for the detection and\ntreatment of endometriosis. 5-ALA follows the heme\npathway and goes into proto-porphyrin IX. They thus\nare using a similar marker with a similar excitation/\nemission as our TPM detection of hemosiderin. How-\never, 5-ALA has several drawbacks. First as an exogen-\nous material it requires a 24-h hospital stay due to its\nnature as a photo-sensitizer [ 19]. 5-ALA is a precursor\nof protoporphyrin IX (PPIX) in the heme pathway. How-\never, PPIX is not in hemosiderin [ 20], thus 5-ALA tar-\ngets the increased degree of angiogenesis and vascular\nendothelial growth factor expression consistent with\nFig. 2 Endogenous Fluorescence of Hemosiderin to Improve Clinical Detection of Endometriosis. Two-Photon Microscopy (TPM), Histology\npositive/negative for endometriosis based on pathologist ’s determination of glands, stroma, and/or hemosiderin\nCabe et al. Translational Medicine Communications             (2019) 4:9 Page 9 of 11\n\nendometriosis. Our techniques present better diagnostic\noptions because they are label free and target physical\ncomponents of endometriosis rather than metabolic\npathways expressed by endometriosis.\nAnother competing technology is Firefly ™ fluorescence\nimaging which uses indocyanine green (ICG) as an ex-\nogenous marker which identifies blood vessels. It has\nbeen used in endometriosis detection as it can identify\nneovascularization associated with endometriosis [ 21].\nBeyond being an exogenous marker, it has several other\nshortcomings in detection. Specifically, as not all endo-\nmetriosis presents as ‘red’ vascularized endometriosis, its\nability to accurately detect black and white endometri-\nosis is poor and other methods that do not rely on\nangiogenesis are needed [ 22].\nA criticism of using TPM to identify hemosiderin is\nthe high cost of the two photon instrument to be used\nin the operating room. As a result, we evaluated whether\nthe low cost option of single photon fluorescence could\nbe implemented to also identify hemosiderin in endo-\nmetriosis and found similar accuracy. Clinical translation\nof these results to a cost-effective device has been built,\nprogramed, and its camera and electronic components\nhave been tested. We are currently repeating our studies\nusing this single photon device which can easily be\ninserted into a laparoscope for clinical translation at low\ncost. The clinical application of a laparoscope using the\nspecific fluoroscopy detailed in this paper could provide\nsurgeons with the ability to more accurately detect and\nremoved endometriosis.\nThere are some limitations to our study. Glands and\nhemosiderin are not present in every endometriosis le-\nsion. Even a TPM imaging system would miss some le-\nsions due to their lack of hemosiderin. However it does\npresent an improved option compared to white light, as\nTPM has better statistical accuracy. Only 18 of 27 pa-\ntients identified by the surgeon as having endometriosis\nhad OCT and TPM imaging performed. The clinical\nstudy was started before we fully solved the difficult\ntechnical issue of co-registration of histology and the re-\ngion of tissue optically imaged. Embedding the tissue in\na cube of agar allowed us to maintain tissue orientation\nand thus co-registration. Finally, our studies only en-\nrolled a limited number of subjects. Our findings will be\nexpanded on a larger number of patients focusing on\nsingle photon in the future. A prospective randomized\ntrial in vivo will have to be completed to ultimately\nprove the relevance of our findings.\nConclusions\nIn summary, focusing on hemosiderin with single pho-\nton fluorescence should yield a surgical device capable\nof improving the detection of endometriosis and allow-\ning clinical translation in a cost effective approach. Both\nglands seen by OCT and hemosiderin seen by TPM im-\nprove a surgeon ’s ability to correctly diagnose endomet-\nriosis. This is demonstrated in improved negative and\npositive predictive values as compared to the clinical im-\npression of the surgeon. The improved diagnosis by\nOCT-TPM may result in fewer false positive results and\nincreased disease tissue removed, although this will have\nto be tested in a prospective clinical trial.\nAbbreviations\n5-ALA: δ-Aminolevulinic acid; ICG: Indocyanine green; NPV: Negative\npredictive values; OCT: Optical coherence tomography; PPIX: Protoporphyrin\nIX; PPV: and positive; TPM: Two photon microscopy\nAcknowledgements\nNot applicable.\nFunding\nClayton Foundation for Biomedical Research, Houston, Texas, USA.\nThe funders had no role in the initiation or design of the study, collection of\nsamples, analysis, interpretation of data, writing of the paper, or the\nsubmission for publication.\nAvailability of data and materials\nThe datasets used and/or analysed during the current study not included in\nthis published article are available from the corresponding author on\nreasonable request.\nAuthors’ contributions\nAGC planned the study, performed all experiments, collected the data,\nanalyzed the results, and wrote the manuscript. ADE helped conduct\nimaging, devised image processing methods, and contributed to study\ndesign. RDR, BC, and JEM preformed surgeries and provided all biopsies. TH\nand XY processed imaging data and assisted in image co-registration. SJ\ncontributed to designing image processes and data acquisition. PTV was the\nexpert pathologist and analyzed all histology slides. TEM, RDR, and MDF con-\ntributed to study design. ADE, JEM, RDR, TH, TE M, and MDF all assisted in re-\nvising the manuscript. All authors read and approved the final version.\nEthics approval and consent to participate\nIRB approval was received and maintained at both sites. Informed consent\nwas received from every patient.\nConsent for publication\nNot applicable.\nCompeting interests\nThe following authors have potential conflicts of interest and have\ncompleted International Committee of Medical Journal Editors (ICMJE) forms\nper their guidelines.\nAGC reports personal fees from Clayton Foundation for Research, during the\nconduct of the study; In addition, AGC has a patent International Patent\nApplication No. PCT/US2018/032877 based on U.S. Serial No. 62/506,910;\nEntitled “Systems and Methods for Endometrial Tissue Identification ” pending\nto Clayton Foundation for Research.\nADE reports personal fees from Clayton Foundation for Research, during the\nconduct of the study; In addition, ADE has a patent International Patent\nApplication No. PCT/US2018/032877 based on U.S. Serial No. 62/506,910;\nEntitled “Systems and Methods for Endometrial Tissue Identification ” pending\nto Clayton Foundation for Research.\nTH reports personal fees from Clayton Foundation for Medical Research,\nduring the conduct of the study.\nBC reports personal fees from Clayton Foundation for Research, during the\nconduct of the study.\nRDR reports grants from AbbVie, outside the submitted work.\nTEM reports personal fees from Clayton Foundation for Research, during the\nconduct of the study; In addition, TEM has a patent International Patent\nApplication No. PCT/US2018/032877 based on U.S. Serial No. 62/506,910;\nCabe et al. Translational Medicine Communications             (2019) 4:9 Page 10 of 11\n\nEntitled “Systems and Methods for Endometrial Tissue Identification ” pending\nto Clayton Foundation for Research.\nMDF reports personal fees from Clayton Foundation for Research, during the\nconduct of the study; In addition, MDF has a patent International Patent\nApplication No. PCT/US2018/032877 based on U.S. Serial No. 62/506,910;\nEntitled “Systems and Methods for Endometrial Tissue Identification ” pending\nto Clayton Foundation for Research.\nAll other authors declare that they have no competing interests.\nPublisher’sN o t e\nSpringer Nature remains neutral with regard to jurisdictional claims in\npublished maps and institutional affiliations.\nAuthor details\n1Departments of Cardiology, University of Texas Health Science Center at San\nAntonio, 7703 Floyd Curl Drive, San Antonio, TX 78229, USA. 2Departments\nof Obstetrics and Gynecology, University of Texas Health Science Center at\nSan Antonio, 7703 Floyd Curl Drive, San Antonio, TX 78229, USA. 3School of\nBiomedical Engineering, Cockrell School of Engineering, The University of\nTexas at Austin, 107 W. Dean Keeton, BME Building, 1 University Station,\nC0800, Austin, TX 78712, USA. 4Departments of Pathology, University of Texas\nHealth Science Center at San Antonio, 7703 Floyd Curl Drive, San Antonio,\nTX 78229, USA. 5Seven Oaks Women ’s Center, San Antonio, Seven Oaks\nWomen’s Center, 7711 Louis Pasteur, Suite 200, San Antonio, TX 78229, USA.\n6Department of Medicine, University of Texas Health Science Center, 7703\nFloyd Curl Drive, San Antonio, TX 78229, USA.\nReceived: 14 February 2019 Accepted: 30 April 2019\nReferences\n1. Jacobson TZ, Duffy JM, Barlow DH, Farquhar C, Koninckx PR, Olive D.\nLaparoscopic surgery for subfertility associated with endometriosis. In:\nJacobson TZ, editor. Cochrane Database of Systematic Reviews. Chichester:\nWiley; 2010. https://doi.org/10.1002/14651858.CD001398.pub2.\n2. Marcoux S, Maheux R, Bérubé S. Endometriosis the CCG onLaparoscopic\nsurgery in infertile women with minimal or mild endometriosis. N Engl J\nMed. 1997;337(4):217 –22. https://doi.org/10.1056/NEJM199707243370401.\n3. Shaw RW. Treatment of endometriosis. Lancet. 1992;340(8830):1267 –71.\nhttps://doi.org/10.1016/0140-6736(92)92960-N.\n4. Stegmann BJ, Sinaii N, Liu S, et al. Using location, color, size, and depth to\ncharacterize and identify endometriosis lesions in a cohort of 133 women.\nFertil Steril. 2008;89(6):1632 –6. https://doi.org/10.1016/j.fertnstert.2007.05.042.\n5 . A b b o t tJ A ,H a w eJ ,C l a y t o nR D ,G a rry R. The effects and effectiveness\nof laparoscopic excision of endometriosis: a prospective study with 2-\n5 year follow-up. Hum Reprod. 2003;18(9):1922 –7. https://doi.org/10.\n1093/humrep/deg275 .\n6. Brosens I, Puttemans P, Campo R, Gordts S, Kinkel K. Diagnosis of\nendometriosis: pelvic endoscopy and imaging techniques. Best Pract\nRes Clin Obstet Gynaecol. 2004;18(2):285 –303. https://doi.org/10.1016/j.\nbpobgyn.2004.03.002 .\n7. Guy CT, Webster MA, Schaller M, Parsons TJ, Cardiff RD, Muller WJ.\nExpression of the neu protooncogene in the mammary epithelium of\ntransgenic mice induces metastatic disease. Proc Natl Acad Sci. 1992;89(22):\n10578–82. https://doi.org/10.1073/pnas.89.22.10578.\n8. Perry SW, Burke RM, Brown EB. Two-photon and second harmonic\nmicroscopy in clinical and translational Cancer research. Ann Biomed Eng.\n2012;40(2):277–91. https://doi.org/10.1007/s10439-012-0512-9.\n9. Boppart SA, Goodman A, Libus J, et al. High resolution imaging of\nendometriosis and ovarian carcinoma with optical coherence tomography:\nfeasibility for laparoscopic-based imaging. BJOG An Int J Obstet Gynaecol.\n1999;106(10):1071–7. https://doi.org/10.1111/j.1471-0528.1999.tb08116.\n10. Brosens I, Gargett CE, Guo S-W, et al. Origins and progression of adolescent\nendometriosis. Reprod Sci. 2016;23(10):1282 –8. https://doi.org/10.1177/\n1933719116637919.\n11. Mettler L, Schollmeyer T, Lehmann-Willenbrock E, et al. Accuracy of\nlaparoscopic diagnosis of endometriosis. JSLS. 2003;7(1):15 –8 http://www.\nncbi.nlm.nih.gov/pubmed/12722993.\n12. Stratton P, Winkel C, Premkumar A, et al. Diagnostic accuracy of\nlaparoscopy, magnetic resonance imaging, and histopathologic examination\nfor the detection of endometriosis. Fertil Steril. 2003;79(5):1078 –85. https://\ndoi.org/10.1016/S0015-0282(03)00155-9.\n13. Stratton P, Winkel CA, Sinaii N, Merino MJ, Zimmer C, Nieman LK. Location,\ncolor, size, depth, and volume may predict endometriosis in lesions\nresected at surgery. In: Fertility and sterility, vol. 78: Elsevier; 2002. p. 743 –9.\nhttps://doi.org/10.1016/S0015-0282(02)03337-X.\n14. Kinkel K, Frei KA, Balleyguier C, Chapron C. Diagnosis of endometriosis with\nimaging: a review. Eur Radiol. 2006;16(2):285 –98. https://doi.org/10.1007/\ns00330-005-2882-y.\n15. Vlek SL, Lier MCI, Ankersmit M, et al. Laparoscopic imaging techniques in\nendometriosis therapy: a systematic review. J Minim Invasive Gynecol. 2016;\n23(6):886–92. https://doi.org/10.1016/j.jmig.2016.06.019.\n16. Guan X, Nguyen MTA, Walsh TM, Kelly B. Robotic single-site endometriosis\nresection using firefly technology. J Minim Invasive Gynecol. 2016;23(1):10 –\n1. https://doi.org/10.1016/j.jmig.2015.08.001.\n17. Demco L. Detection of endometriosis currently depends on the\nvisualization and identification of peritoneal lesions at the time of pelvic\nlaparoscopy. With improved optics, smaller lesions became visible and\ndescribed. J Am Assoc Gynecol Laparosc. 2004;11(2):2002 –5.\n18. Buchweitz O, Staebler A, Tio J, Kiesel L. Detection of peritoneal\nendometriotic lesions by autofluorescence laparoscopy. Am J Obstet\nGynecol. 2006;195(4):949 –54. https://doi.org/10.1016/j.ajog.2006.02.044.\n19. Hillemanns P, Weingandt H, Stepp H, Baumgartner R, Xiang W, Korell M.\nAssessment of 5-aminolevulinic acid –induced porphyrin fluorescence in\npatients with peritoneal endometriosis. Am J Obstet Gynecol. 2000;183(1):\n52–7. https://doi.org/10.1067/mob.2000.105897.\n20. Beaumont C, Delaby C. Recycling Iron in Normal and pathological states.\nSemin Hematol. 2009;46(4):328 –38. https://doi.org/10.1053/j.seminhematol.\n2009.06.004.\n21. Cosentino F, Vizzielli G, Turco LC, et al. Near-infrared imaging with\nIndocyanine green for detection of endometriosis lesions (Gre-Endo trial): a\npilot study. J Minim Invasive Gynecol. March 2018. https://doi.org/10.1016/J.\nJMIG.2018.02.023.\n22. Laschke MW, Menger MD. In vitro and in vivo approaches to study\nangiogenesis in the pathophysiology and therapy of endometriosis. Hum\nReprod Update. 2007;13(4):331 –42. https://doi.org/10.1093/humupd/\ndmm006.2016.06.019.\nCabe et al. Translational Medicine Communications             (2019) 4:9 Page 11 of 11","source_license":"CC0","license_restricted":false}