Detection of cervical intraepithelial neoplasia in vivo using confocal endomicroscopy.

OBJECTIVE: A high resolution optical imaging device may offer a clinically useful adjunct to colposcopy for the diagnosis and assessment of cervical precancer. This study describes the clinical evaluation of a miniaturised confocal endomicroscope for the quantitative and qualitative assessment of cervical intraepithelial neoplasia (CIN) in vivo. DESIGN: A descriptive study (n = 25) was performed to assess the usability of confocal endomicroscopy to image the cervix. A prospective study (n = 15) then evaluated the diagnostic accuracy of the technique. SETTING AND POPULATION: Patients undergoing colposcopy for treatment of CIN1-CIN3 were examined using confocal endomicroscopy. METHODS: A 5% solution of acetic acid was used to enhance the colposcopic features of the atypical region. Normal and abnormal regions of the cervix were then imaged following topical application of a fluorescent dye (acriflavine). MAIN OUTCOME MEASURES: Confocal images were analysed to develop a scoring system to grade different levels of CIN. Microscopic features were correlated with histology from biopsy. RESULTS: Confocal endomicroscopy enabled microscopic imaging of cellular and subcellular structures in vivo at colposcopy. Imaging at increasing depth showed morphological features including dermal papillae, endocervical glands and the squamo-columnar junction. CIN was characterised by an increase in nuclear density, size and cellular atypia. The sensitivity for detection of CIN was 97%. The specificity for predicting the grade of abnormality was 80% for normal-CIN1 and 93% for CIN2-CIN3. CONCLUSIONS: Confocal endomicroscopy is a sensitive imaging tool for detection and assessment of CIN. The technique enables in vivo imaging of cervical histology and the potential for 'see-and-treat' workflows.

In vivo fibreoptic confocal imaging (FOCI) of the human ocular surface.

Recent developments in the miniaturization of confocal imaging technology have resulted in the development of a hand-held confocal microscope probe. There are many structures of interest in the human eye that are within reach of a fluorescence-mode confocal microscope; this study assessed the feasibility of in vivo human ocular imaging. Safety analysis was undertaken to ensure that the laser light applied to the ocular surface structures constituted no threat to patient safety. A fibreoptic confocal imaging (FOCI) probe using an illumination wavelength of 488 nm was applied to the ocular surface of four volunteers after topical administration of sodium fluorescein. Stabilization of the probe on the ocular surface was difficult, but movement artefacts could be minimized to a satisfactory level in most subjects by a variety of procedures. High-quality images of conjunctival epithelial and goblet cells, lamina propria structures, accessory lacrimal glands, lacrimal ducts and superficial sclera were obtained. Lateral resolution was 1-1.5 microm and axial resolution was approximately 30 microm; individual erythrocytes could be seen in conjunctival vessels. The rete ridges and intervening epithelial components, including the probable location of corneal limbal stem cells, could be viewed, although it was not possible to distinguish cell subgroups. The study showed that fluorescence-mode imaging of the ocular surface is a viable and promising tool for assessment of diseases and processes involving superficial ocular structures. Refinement of equipment and techniques, particularly probe stabilization, is necessary to realize fully the potential of FOCI for ocular use.

Autofluorescence of skin burns detected by fiber-optic confocal imaging: evidence that cool water treatment limits progressive thermal damage in anesthetized hairless mice.

BACKGROUND: Although full-thickness burns present no difficulty to clinical judgment, accurate assessment of burn depth immediately after injury in partial thickness burns has always been difficult. METHODS: Thermal burns (applied by a 3-mm-diameter brass rod heated to 50 degrees--80 degrees C for 20 seconds) were induced on the skin of anesthetized hairless mice. Anesthesia was maintained throughout all experiments. Both burns and normal skin were investigated noninvasively in vivo using fiber-optic confocal imaging (FOCI) microscopy (excitation, 488 nm; detection, 505 nm). RESULTS: Autofluorescence was detected in burned skin, and the depth of the autofluorescent region was found to correlate with the intensity of heat applied. Cool water treatment (for 20 minutes immediately after burn induction) significantly reduced the progressive increase in autofluorescence in deeper layers of the skin over the 4-hour postburn observation period. Histology showed burn-associated changes at a lower temperature than that at which autofluorescence was first detected in vivo by FOCI. However, there was a good correlation (r = 0.78) between depth of damage revealed by FOCI compared with that by histology. CONCLUSION: These results suggest that FOCI may be used to provide an index of burn depth.

In vivo detection of small subsurface melanomas in athymic mice using noninvasive fiber optic confocal imaging.

Fiber optic confocal imaging, following intravenous administration of fluorescently labeled antibodies and Texas Red-dextran, enabled in vivo detection of melanoma and surrounding blood vessels in athymic mice. Human melanoma cells (three cell lines) and cultured normal human skin cells were implanted intradermally into the haunch skin of anesthetized athymic BALB/C mice and allowed to grow to a maximum size of 2 mm diameter. Using three different fluorescein-isothiocyanate-labeled antimelanoma antibodies, single channel confocal images of melanoma cells were obtained in vivo. Using noninvasive techniques, the overall in vivo melanoma detection rate for tumors within 0.2 mm of the skin surface was 84% (27 of 32 tumors). Normal cultured human skin cells were found to have little or no fluorescence after administration of the fluorescein-isothiocyanate-labeled antibodies and tumors were not labeled by an isotype control antibody. Dual channel imaging of the implanted melanoma tumor and surrounding dermal vasculature in vivo showed increased blood vessel density at the melanoma site. Conventional immunoperoxidase histology confirmed that fiber optic confocal imaging was able to detect melanoma tumors up to 0.2 mm below the skin surface, in vivo.

A novel, rapid, computerized method for quantitation of neuronal damage in a rat model of stroke.

Determination of extent of infarction in animal models of cerebral ischemia is most commonly achieved by either classical histology (thionin staining) and light microscopy or staining with 2,3, 5-triphenyltetrazolium chloride (TTC). These techniques have limitations and we now describe a novel technique and its validation for assessment of the neuroprotective activity of AM-36, a novel arylalkypiperazine compound with combined antioxidant and sodium channel blocking activity. AM-36 (1.8 mg/kg i.p.) or vehicle, was administered 30 min, 24 and 48 h after endothelin-1-induced middle cerebral artery occlusion in conscious rats. Rats were killed at 72 h, brains removed and frozen in liquid nitrogen prior to coronal sectioning. Using a simple apparatus relying on basic principles of light propagation and a computerised image analysis system, ischemic damage in unstained slide-mounted sections was clearly visualised and measured. AM-36 significantly reduced the area of infarct in both cortex and striatum. The method was verified by thionin staining, and light microscopy. Linear regression analysis showed a highly significant correlation between methods at 72 h for infarct area in the cortex and striatum. Highly significant correlations between methods were found at 3 and 24 h after ischemia. Our method quickly and clearly delineates areas of damage in a manner superior to conventional staining methods.

In vivo mapping of the vascular changes in skin burns of anaesthetised mice by fibre optic confocal imaging (FOCI).

Burns (3 mm in diameter, 50 degrees C, 20 s duration) were induced on the skin of anaesthetised hairless mice. Anaesthesia was maintained throughout all experiments. Subsurface changes in the microvasculature at the burn site were imaged confocally following i. v. injection of fluorescently labelled (FITC) dextran. Blood cells moving through dermal blood vessels were seen and recorded on video tape. Multiple adjacent 2-D confocal images of the burn site and surrounding areas were assembled and enabled microscopic vascular imaging of the whole burn area (including zones of coagulation, stasis and hyperaemia) and the surrounding normal vessels. This mapping of the burn area by fibre optic confocal imaging (FOCI) in vivo demonstrated good congruence with vascular casts (Microfil MV-120, Flow tech, USA) made at 4 h post burn. This study demonstrates the usefulness of FOCI for in vivo vascular imaging in burns.

Live confocal microscopy of oligonucleotide uptake by keratinocytes in human skin grafts on nude mice.

Anti-sense oligonucleotide uptake by keratinocytes in human skin grafts on athymic mice was examined using live confocal microscopy. Fluorescein isothiocyanate-labeled 15-mer C-5 propyne modified phosphorothioate anti-sense oligonucleotide (10-50 microM) was intradermally injected into normal human skin grafts on athymic mice, and the localization of the anti-sense oligonucleotide was assessed after 1-24 h postinjection. Anti-sense oligonucleotide was found to localize in the nuclei of basal and suprabasal keratinocytes after 1-2 h, and this localization was still observed after 24 h. This live in vivo observation of anti-sense oligonucleotide uptake in basal keratinocytes was confirmed using conventional fluorescence microscopy of fixed sections of skin grafts. Neither single nucleotides which were fluorescein isothiocyanate-labeled nor fluorescein isothiocyanate alone was able to penetrate into the nuclei of human skin graft keratinocytes after intradermal injection, and hence it is likely that the anti-sense oligonucleotide was not degraded prior to intracellular localization. Topical administration of anti-sense oligonucleotide and anti-sense oligonucleotide-liposome complexes resulted primarily in localization in the stratum corneum of human skin grafts. When grafts were tape stripped prior to anti-sense oligonucleotide administration, however, as little as 5 microM anti-sense oligonucleotide was required to observe nuclear anti-sense oligonucleotide accumulation. These results suggest that cutaneous anti-sense strategies can be tested using delivery via intradermal anti-sense oligonucleotide injection in human skin grafts on athymic mice, and that agents providing penetration of anti-sense oligonucleotide across the stratum corneum are likely to be required for successful topical therapies.

In vivo fibre optic confocal imaging of microvasculature and nerves in the rat vas deferens and colon.

A fluorescence confocal microscopy technique was employed to obtain subsurface images of nerve and microvascular structure in the vas deferens and colon of the living rat. The use of dual labelling with vital dyes and 2-channel confocal acquisition allowed differentiation of microscopic structure at both low and higher magnification. Characteristic staining patterns of nerves and blood vessels were repeatedly obtained in each tissue, suggesting the potential of this technique for studying morphological changes associated with surgical procedures and/or models of neuronal or vascular pathology.

A study of vascular response to thermal injury on hairless mice by fibre optic confocal imaging, laser doppler flowmetry and conventional histology.

Burn injury causes vascular thrombosis and occlusion by thermal damage to the vascular network in the dermis. In this study, fibre optic confocal imaging (FOCI) and laser doppler flowmetry were used to detect changes in vascular morphology and local dermal blood flux over 4 h, in three defined zones after a thermal burn (50 degrees C, 20 s duration, 3 mm in diameter) was induced on fully anaesthetised hairless mice. FITC-dextran (i.v.) was used to enable FOCI of vascular morphology including three-dimensional imaging of the burn site and its surrounding areas. Samples of the affected areas were collected for conventional histology, including Masson's trichrome. There was vascular damage in the zone of coagulation which showed no change during the 4 h period. The zone of stasis showed an initial reduction in blood flux and confocal imaging of the area indicated significant vessel leakage during the first 2 h which later improved. The zone of hyperaemia showed an initial increase in total blood flux and confocal imaging of the area showed initial blood vessel dilatation. This study demonstrates that FOCI is a useful non-invasive tool in the assessment of vascular changes in thermal burns in vivo, and compares the findings of FOCI with those from laser doppler flowmetry and histology.

Fibre optic confocal imaging (FOCI) of keratinocytes, blood vessels and nerves in hairless mouse skin in vivo.

Fibre optic confocal imaging (FOCI) enabled subsurface fluorescence microscopy of the skin of hairless mice in vivo. Application of acridine orange enabled imaging of the layers of the epidermis. The corneocytes of the stratum corneum, the keratinocytes in the basal layers and redundant hair follicles were visualised at depths greater than 100 microm. Cellular and nuclear membranes of keratinocytes of the skin were visualised by the use of acridine orange and DIOC5(3). Imaging of the skin after injection of FITC-dextran revealed an extensive network of blood vessels with a size range up to 20 microm. Blood cells could be seen moving through dermal vessels and the blood circulation through the dermal vascular bed was video-taped. The fluorescent dye 4-di-2-ASP showed the presence of nerves fibres around the hair follicles and subsurface blood vessels. Comparison was made between images obtained in vivo using FOCI and in vitro scanning electron microscopy and conventional histology. FOCI offers the potential to study dynamic events in vivo, such as blood flow, skin growth, nerve regeneration and many pathological processes, in ways which have not previously been possible.

Rotary manipulation for cervical radiculopathy: observations on the importance of the direction of the thrust.

OBJECTIVE: To describe the clinical presentation of eight patients with cervical spine radiculopathy, the manipulation technique used for each patient and the outcomes of treatment. CLINICAL FEATURES: The cause of radiculopathy in four patients was disc herniation. The other four patients had a combination of spondylosis, disc herniation and sprain injury. INTERVENTION AND OUTCOME: Six of eight patients had a good outcome associated with receiving manipulation performed by contacting the cervical spine at the level of the radiculopathy, laterally flexing toward the side of radiculopathy, rotating the neck away from the side of the radiculopathy and applying a gentle high-velocity, low-amplitude thrust. Two patients had an exacerbation of arm pain and increased neurological deficit associated with manipulation performed with the neck rotated toward the side of radiculopathy. CONCLUSION: There is little compelling evidence supporting or disputing the use of manipulation for patients with cervical spine radiculopathy. In our patients, rotary manipulation was associated with a different outcome depending on the direction of neck rotation. Prospective time-series studies and randomized, blind trials are needed to identify the efficacy and effectiveness of different manipulation techniques for this condition.

Fibre optic confocal imaging (FOCI) for subsurface microscopy of the colon in vivo.

Fibre optic confocal imaging (FOCI) is a new type of microscopy which has been recently developed (Delaney et al. 1993). In contrast to conventional light microscopy, FOCI and other confocal techniques allow clear imaging of subsurface structures within translucent objects. However, unlike conventional confocal microscopes which are bulky (because of a need for accurate alignment of large components) FOCI allows the imaging end to be miniaturised and relatively mobile. FOCI is thus particularly suited for clear subsurface imaging of structures within living animals or subjects. The aim of the present study was to assess the suitability of using FOCI for imaging of subsurface structures within the colon, both in vitro (human and rat biopsies) and in vivo (in rats). Images were obtained in fluorescence mode (excitation 488 nm, detection above 515 nm) following topical application of fluorescein. By this technique the glandular structure of the colon was imaged. FOCI is thus suitable for subsurface imaging of the colon in vivo.

Fiber-optic laser scanning confocal microscope suitable for fluorescence imaging.

A fiber-optic laser scanning confocal microscope that utilizes a single-mode optical fiber as both the source and the detection aperture is described. The design of a compact, portable scanning head that is robust in alignment and fits simply to standard light microscopes is presented. Resolution performance and imaging capability are demonstrated for several objective lenses in both reflectance and fluorescence modes of operation. This microscope achieves performance that is comparable with that of conventional bulk optical laser scanning confocal microscopes while offering flexibility in usage forbidden to such instruments.