1207 Laser Physics, Vol. 14, No. 9, 2004, pp. 1207–1213. Original Text Copyright © 2004 by Astro, Ltd. English Translation Copyright © 2004 by MAIK “Nauka /Interperiodica” (Russia). INTRODUCTION The existing extensive experience in diagnosing and treating oncological diseases makes it possible to assess the importance and complexity of the timely diagnostics of cancer and precursors of cancer. The existing means for intraoperative diagnostics in gyne- cology (visual examination, including endoscopic study, palpation, and urgent histological study) are insufficient for surgeons, since urgent histological or cythological studies, as well as visual and palpatory examinations, often yield erroneous results, leading to inadequate decisions at important diagnostic stages. Thus, the importance of early and precise diagnostics in gynecology has stimulated a search for fundamentally new diagnostic procedures. In the development of new diagnostic tools, attention is mostly given to optical methods [1]. Fluorescence diagnostics (FD) involves a spectral analysis of living tissues. Optical diagnostics is based on a selective excitation of fluorescence in tissues using a laser or an alternative light source with a certain wavelength. One can use both endogenous and exoge- nous photosensitizers. The selectivity of their distribu- tion in the body is related to biochemical and physio- logical differences between tissues and the properties of the intercellular medium. Spectroscopic analysis of fluorescence signals makes it possible to distinguish between healthy and pathological tissues. Fast data pro- cessing enables one to make corrections to the treat- ment [2]. One of the topical problems in FD is the search for optimal photosensitizers. δ -5-Aminolevulinic acid (5-ALA) and the Alasens preparation based on 5-ALA cannot be classified as photosensitizers. However, these substances induce the intracellular synthesis of proto- porphyrin IX (PP IX), which exhibits high-intensity fluorescence and photodynamic activity. The high accumulation of PP IX in rapidly proliferating cells and its rapid utilization in normal cells serve as the biolog- ical basis for FD. The aforesaid processes account for a high fluorescence contrast of the pathological tissue vs. the normal tissue. Rapid production of PP IX and its rapid utilization in normal cells yielding photoinactive heme (this accounts for the absence of phototoxicity) have led to extensive clinical study of FD using 5-ALA [3–7]. It is known that blue and green light used for fluo- rescence excitation and fluorescence imaging are readily absorbed by blood hemoglobin. This leads to a decrease (down to 0.5 mm) in the measurement depth in the tissues under study [8] and causes errors in esti- mates of the real concentration of PP IX owing to the variation in the hemoglobin content in tissues [9]. The aforesaid facts necessitate spectral analysis of the fluo- rescence. We measure fluorescence spectra using red laser excitation with a wavelength of 633 nm, so the depth of optical biopsy of the tissues is about 1–3 mm. To increase the efficiency of early and differential diagnostics of gynecological diseases, we develop an FD method that can be used in gynecological opera- BIOPHOTONICS Laser Fluorescence Spectroscopy with 5-Aminolevulinic Acid in Operative Gynecology L. A. Belyaeva 1, 2 , L. V. Adamyan 1 , A. A. Stepanyan 1 , K. G. Linkov 2 , and V. B. Loshchenov 2 1 Department of Reproductive Medicine and Surgery, Faculty of Postgraduate Education, Moscow State University of Medicine and Dentistry, Moscow, Russia 2 Natural Sciences Center, Prokhorov General Physics Institute, Russian Academy of Sciences, ul. Vavilova 38, Moscow, 119991 Russia e-mail: l_b@bk.ru Received September 10, 2003 Abstract —A method for the fluorescence intraoperative diagnostics of gynecological diseases using a laser– fiber spectrum analyzer and Alasens-induced protoporphyrin IX (PP IX) is developed, and the efficiency of the method is estimated. In the spectroscopic study, the fluorescence of PP IX is excited using laser radiation with a wavelength of 633 nm and is measured during laparoscopy, laparotomy, hysteroscopy, physition examination, and ex vivo in 75 patients with genital endometriosis and various pathologies of the ovary, vulva, and cervix and body of the uterus. The spectroscopic results are compared to histological diagnosis. The data obtained show laser fluorescence spectroscopy using Alasens to be very effect in differential diagnostics of ovary diseases, cer- vical and vulvar cancer, and metastases. Additional study is needed to estimate the applicability of the method under consideration for intraoperative diagnostics of active peritoneal endometriosis. A target biopsy of visually intact fragments can be realized with the aid of fluorescence diagnostics based on laser fluorescence spectros- copy of the Alasens-induced PP IX in organs and tissues. High accumulation of PP IX in normal and patholog- ical endometrium impedes application of the fluorescence method for purposes of differential diagnostics but opens up prospects for photodynamic ablation of endometrium. 1208 LASER PHYSICS Vol. 14 No. 9 2004 BELYAEVA et al . tions and that is based on laser spectroscopic measure- ments of the concentration of Alasens-induced PP IX in tissues. For this purpose, we carry out a spectroscopic study, estimate the fluorescence spectral properties of tissues in the organs of female genitals, and compare the results obtained with histological data. We study normal tissues and tissues from patients with endometriosis ovarial, endometrial, cervical, and vul- var benign and malignant tumors. We develop new methods to measure fluorescence in order to optimize the FD procedures. MATERIALS AND METHODS We perform fluorescence intraoperative measure- ments in 75 gynecological patients in the course of lap- arotomy, laparoscopy, hysteroscopy, and physition examination. The age of the patients ranges from 20 to 77 years. Below, we present detailed information on the patients and demonstrate examples and the results obtained. All patients undergo conventional a scheduled exam- ination prior to operation. None of the patients with por- phyria. During operations, FD supplements conven- tional diagnostic methods. We analyze spectroscopic characteristics of tissues in real time. The mean time of optical express diagnostics of tissues is 2–3 min. We employ 5-ALA; the commercial name of the preparation produced by NIOPIK (Moscow, Russia) is Alasens. Equipment For the intraoperative FD of gynecological diseases, we employ conventional surgical instruments and equipment and an LESA-01-Biospec electronic laser spectral setup. The main components of this setup are shown in Fig.1 and comprise (i) a helium–neon laser with a wavelength of 632.8 nm and an output power of 1–10 mW, (ii) a fiber-optic multichannel catheter as a system consisting of two parts (the laser system is used to deliver the laser radiation to the object under study, and the detecting and measuring system is used to detect fluorescence and scattered laser light and to transmit it to the detector), (iii) a set of color filters with a transmission band of 630–750 nm, (iv) a multichannel spectrum analyzer allowing for fast spectral measure- ments, and (v) a PC. The data obtained are processed using original computer codes. Method for Intraoperative FD Patients were informed about the procedure. After obtaining the patient’s consent, we orally introduce about 40 ml of Alasens at a doze of 25 mg/kg. The diag- nostic procedures are started from three to eight hours after the preparation is introduced. During gynecologi- cal operations, we perform the fluorescence spectral analysis as follows. A diagnostic fiber (with a diameter of 1.8 mm) delivering radiation from a helium–neon laser is brought in light contact with the tissue under study. In the case of laparotomy and external examina- tion, we do this directly by hand. In the case of laparos- copy, we employ a additional troacar and an aspirating needle. In the case of hysteroscopy, we introduce the fiber via the operation channel. For spectral measure- Fig. 1. LESA-01-Biospek spectroscopic system. LASER PHYSICS Vol. 14 No. 9 2004 LASER FLUORESCENCE SPECTROSCOPY 1209 ments, we need to switch off the source of light for a short period of time. For comparative analysis, we per- form a preoperative measurement of the accumulation of the Alasens-induced PP IX in the skin of the internal surface of the forearm and in the mucous of the patient’s lip and measure the autofluorescence of the healthy skin. We also perform a spectral analysis of the removed tissues immediately after operation. Algorithm to Process FD Results FD results are given in fluorescence spectra, in which we plot the wavelength in nanometers and the intensity of fluorescence and scattered laser radiation in arbitrary units on the x and y axes, respectively. The sharp peak at 633 nm corresponds to the scattered laser light, whereas the fluorescence is seen as a broad band. We analyze the spectral shapes and signal amplitudes. To construct histograms, we calculate the fluorescence coefficient: (1) We use the fluorescence coefficient as a diagnostic criterion for the PP IX relative concentration in tissues. The values of k f are normalized by the value of the coef- ficient corresponding to normal skin of the arm. To estimate the difference between the fluorescence intensities of normal and pathological tissues, we employ the diagnostic contrast coefficient: (2) k f Area_under_Fluorescense_curve Area_under_Laser_scattered_peak .= k dc k f pathol() k f norma() .= We analyze the laser fluorescence spectral data for 2000 points of the tissues under study (20–40 points per patient). RESULTS AND DISCUSSION To estimate the efficiency of laser fluorescence spectroscopy, we employ such parameters as sensitivity and specificity. The sensitivity shows if FD can be used for diagnosing a certain pathology. The specificity characterizes the ability of the method to rule out the possibility of the pathology. Normal Tissues The processing of spectral data yields the minimum accumulation of PP IX in normal peritoneum; serous tissue of the uterus, uterine tube, and myomas; and in normal ovary tissues ( k f = 2.1–3.8). Based on these data, we can develop a method for panoramic intraperi- toneal fluorescence visualization that enables one to compare normal tissue with pathological tissue exhibit- ing anomalously high-intensity fluorescence. We find a high accumulation of the Alasens-induced PP IX in the fimbrii. The spectral analysis shows that the PP IX accumulation in normal endometrium ( k f = 11.6) is higher than that in myometrium ( k f = 3.6) by a factor of 3.5. Ovarian Diseases Benign ovarian pathologies ( n = 22) (follicular cyst, corpus luteum cyst, endometrioid cyst, fibrothecoma, [0] h-n [2] lips-p [3] big t-r ovarii [9] periton-n 3600 3400 3200 3000 2800 2600 2400 2200 2000 1800 1600 1400 1200 1000 800 600 400 200 600 650 700 750 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 0 2 3 9 Fig. 2. Patient M. aged 40 years. Diagnosis: corpus luteum cyst with hemorrhage. 1210 LASER PHYSICS Vol. 14 No. 9 2004 BELYAEVA et al . benign ovarian cystadenomas) are characterized by a low intensity of the PP IX fluorescence ( k f = 3). In the case of endometrioid cysts and cysts with hemorrhage, we occasionally observe high values of k f related to the contribution of autofluorescence (this is clearly seen from the spectral shapes in Fig. 2). For the borderline ovarian tumors ( n = 3), the fluo- rescence coefficient is k f = 29.5 and the diagnostic con- trast coefficient with respect to the normal ovarian tis- sue is k dc = 9 (Fig. 3). It is demonstrated that a high fluorescence intensity of the Alasens-induced PP IX is typical of ovarian can- 2000 620 1800 1600 1400 1200 1000 800 600 400 200 640 660 680 700 720 740 760 780 [0] h-n [5] c-r-ovar [7] ovar 800 2200 2400 2500 2600 3000 [10] tub-ser [11] tub-met-ser [16] n-serosa-uter [17] serosa-met [24] periton-n [25] perit-met [27] met-salnk 0 5 10 15 20 25 30 25 0 5 7 10 11 16 17 24 25 27 Fig. 4. Patient D. aged 56 years. Diagnosis: T 3 c N 0 M 1 ovarian cancer. 2000 620 1800 1600 1400 1200 1000 800 600 400 200 640 660 680 700 720 740 760 780 [0] h-n [3] ovar-n-left [10] kistoma-right 0 2 4 6 8 10 12 14 16 18 20 22 24 26 0 3 10 Fig. 3. Patient M. aged 45 years. Diagnosis: ovarian serous cystadenoma of borderline malignancy. LASER PHYSICS Vol. 14 No. 9 2004 LASER FLUORESCENCE SPECTROSCOPY 1211 cer ( n = 9): k f = 45.6 and k dc = 17.2 (Fig. 4). We also demonstrate statistically valid differences between the results obtained for ovarian metastases and the normal ovarian tissues. We can also reliably distinguish between the metastases of lymph nodes, peritoneum, and greater omentum, on the one hand, and the normal unchanged tissues, on the other hand. The method under consideration demonstrated 100% sensitivity and specificity in the differential diagnostics of cysto- mas with cancer and ovarian metastases with border- line cystadenomas versus normal ovary and benign cysts. [0] h-n [13] endocerv-suspicion [16] ectocerv-suspicion 2800 2600 2400 2200 2000 1800 1600 1400 1200 1000 800 600 400 200 600 650 700 750 0 13 17 5 10 15 20 25 30 35 0 [17] endocer [18] ectocer 16 18 Fig. 6. Patient K. aged 55 years. Diagnosis: T 1 A 1 N 0 M 1 cervical cancer. [0] h-n [6] endom-n [3] endom-left-and 3600 3400 3200 3000 2800 2600 2400 2200 2000 1800 1600 1400 1200 1000 800 600 400 200 600 650 700 750 0 6 7 5 10 15 20 25 30 35 0 Fig. 5. Patient Yag. aged 44 years. Diagnosis: T 1 N 0 M 0 adenocarcinoma. For this patient, the scheduled histological examination revealed microfocuses of high-grade differentiated adenocarcinoma without invasions in myometrium (predominantly, in the left uterine corner). 1212 LASER PHYSICS Vol. 14 No. 9 2004 BELYAEVA et al . Endometrium Pathologies For endometrial cancer ( n = 11), the mean fluores- cence coefficient is k f = 37.3. For the local form of byendometrial cancer, the diagnostic contrast coeffi- cient is k dc = 5.9 (Fig. 3). For the atypical hyperplasia of endometrium ( n = 2), the mean fluorescence coefficient is k f = 31.4. We can- not determine the diagnostic contrast coefficient owing to the diffuse form of the pathology. In the case of endometrium polyp ( n = 9), we also observe a relatively high intensity of the Alasens- induced fluorescence ( k f = 26.4). However, the diagnos- tic contrast coefficient (polyp vs. normal endometrium) is k dc = 2.7. Uterine sarcoma ( n = 1) is characterized by a rela- tively low fluorescence intensity of the Alasens- induced PP IX. In patients with first-stage uterine cancer, cancer in situ , and third-stage dysplasia ( n = 13), we find that the fluorescence intensity of pathological tissues is greater than that of normal tissues by a factor of 5.5 (on the average) (Fig. 6). Figure 7 shows the dependence of the diagnostic contrast coefficient (with respect to intact tissue) of the Alasens-induced PP IX on the stage of the pathological process. Note that the values above the level of 5.3 are related to T 1A, B invasive cervical cancer. The sensitivity and specificity of the method in the differential diag- nostics of cervical cancer versus normal epithelium are estimated as 93 and 99%, respectively. For cancer of the vulva (n = 4), the mean diagnostic contrast coefficient (with respect to normal tissue) is k dc = 5.23 (Fig. 8). The spectroscopic data allow us to differentiate between cancer of the vulva and normal tissue with a sensitivity and a specificity of 100%. For endometriosis of the peritoneum, uterine tubes, and ovary (n = 11), the diagnostic contrast coefficients (with respect to intact tissue) are k dc = 5.2, 3.8, and 4.0, respectively. The diagnostic contrast coefficient can 2000 620 1800 1600 1400 1200 1000 800 600 400 200 640 660 680 700 720 740 760 780 [8] vulva-t-r-left [12] vulva-n-right [18] met-i-u-left 0 2 4 6 8 10 12 14 16 18 20 22 24 26 8 12 18 0 600 28 21 26 [21] l-u-n-right [26] h-n Fig. 8. Patient T. aged 66 years. Diagnosis: T 3 N 1 M 0 cancer of vulva. 9 8 7 6 5 4 3 2 k dc 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 Number of patient Fig. 7. Plot of k dc vs. the stage of cervical pathology: n = 1, CIN I; n = 2, CIN III; n = 3–8, cancer in situ; and n = 9–14, T 1A, B cervical cancer. LASER PHYSICS Vol. 14 No. 9 2004 LASER FLUORESCENCE SPECTROSCOPY 1213 also take a value of 1.6, which corresponds to the absence of reliable difference. A possible reason for this is the proliferative activity of endometriosis. Leiomyoma on the side of the serous tissue of the uterus is characterized by a relatively low k f . In the case of submucosa localization of nodes and, probably, in the case of an actively growing tumor, the fluorescence coefficient can be increased by a factor of 5.4. Alasens is tolerated well by the majority of patients. In a few cases, it causes nausea. At a dose of 30 mg/kg, one patient exhibited hyperemia of the face and chill, which appeared four hours after ingestion and lasted about one day with the introduction of antihistamines. To optimize the FD methods in the diagnostics of pathologies, we develop a new procedure for the fluo- rescence measurement of tissue microsamples (biop- sies, samples of diagnostic curettage, puncture mate- rial). It is difficult to measure the fluorescence spectra of the aforesaid samples owing to the smallness of the volume under study. A direct concentration measure- ment at an object plate yields a relatively high error owing to the losses of laser radiation (part of the laser beam passes by the sample under study, and part is reflected from the glass surface). For the spectroscopic measurements of small volumes of biological tissues, we employ specially designed object plates with an absorbent coating. The sample under study is placed in a glass container with a volume of 1 mm 3 and is covered with a cover glass. This makes it possible to substan- tially improve the results obtained. The results of test measurements show that the above procedure yields an error of 10–20% with respect to the results from fluorescence spectrometry of the tis- sues, whereas in the absence of the special substrate, the relative error of measurements is 100–150%. To measure lateral surfaces (e.g., the mucous tissue of the cervical channel), we employ specially designed heads mounted on the optical fiber. These heads make it possible to take measurements at an angle of 70° rel- ative to the axis. CONCLUSIONS In the course of surgical operations, optical express biopsy makes it possible to differentiate between, on the one hand, benign cysts and cystomas and, on the other hand, ovarian cancer and borderline ovarian tumors and to diagnose micrometastases in the ovary, peritoneum, and omentum. In the cases of endometrial, cervical, vulva cancer, the diagnostic contrast coeffi- cients are determined. It is expedient to develop fluo- rescence diagnostics for screening programs aimed at early diagnosis of premalignancies of the cervix and vulva. Apparently, the differences in PP IX accumula- tion in endometrioid focuses of peritoneum, uterine tubes, and ovary are related to the proliferative activity of endometriosis. A combination of spectrometry and fluorescence imaging will make it possible to diagnose pathologies that are invisible to the eye, to estimate the proliferation of a pathological process, and to monitor tissues after treatment. The results on a relatively high accumulation of the Alasens-induced PP IX in endometrium and a relatively low concentration of PP IX in myometrium are evidence of the promising character of the photodynamic ablation of endometrium. REFERENCES 1. N. Rammanujam, Encyclopedia of Analytical Chemistry (Wiley, Chichester, 2000). 2. R. Baumgartner, Proc. SPIE 3563, 90 (1999). 3. C. J. Kelty, N. J. Brown, M. R. A. Reed, and R. Ackroyd, Photochem. Photobiol. Sci. 1, 158 (2002). 4. M. K. Fehr, P. Wyss, Y. Tadir, et al., Photomedicine in Gynecology and Reproduction (Basel, Karger, 2000). 5. E. Malik, A. Meyhofer-Malik, C. Berg, et al., Hum. Reprod. 15, 584 (2000). 6. P. Hillemanns, H. Weingandt, H. Stepp, et al., Am. J. Obstet. Gynecol. 184 (5), 1046 (2001). 7. L. A. Belyaeva, L. V. Adamyan, V. B. Laschenov, et al., Fluorescence diagnostics in oncological gynecology, Proc. of SPIE vol. 5068 (2003), Saratov Fall Meeting, 2002; Optical Technologies in Biophysics and Medicine, ed. V. V. Tuchin (SPIE, Bellingham, WA, 2003), pp. 55– 60. 8. P. Wyss, L. O. Svaasand, Y. Tadir, et al., Hum. Reprod. 10 (1), 221 (1995). 9. B. Loschenov, V. I. Konov, and A. M. Prokhorov, Laser Phys. 10, 1188 (2000). . method that can be used in gynecological opera- BIOPHOTONICS Laser Fluorescence Spectroscopy with 5-Aminolevulinic Acid in Operative Gynecology L. A and precursors of cancer. The existing means for intraoperative diagnostics in gyne- cology (visual examination, including endoscopic study, palpation, and