The bipedicled osteocutaneous scapula flap: a new subscapular system free flap

Thirty-six adult dissections (14 cadaver and 22 operative) demonstrate the constant presence of the angular branch of the thoracodorsal artery as a vascular pedicle to the inferior pole of the scapula. This vessel originated in all cases just proximal or distal to the serratus branch of the thoracodorsal artery and arborized to the periosteum 6 to 9 cm from the bony branch of the circumflex scapular artery. In eight patients, scapular osteocutaneous flaps were raised preserving the angular branch and the circumflex scapular artery and dissecting up to the subscapular vessels. In all cases, bone was independently perfused by the angular branch. In all six cases where the angular branch was the sole supply to bone, technetium-99m scans demonstrated perfusion. Addition of this vascular pedicle to scapula bone allows two separate bone flaps with one microanastomosis and provides a longer arc of rotation between skin supplied by the circumflex scapular artery and bone. Donor-site morbidity was no greater than with the standard scapula flap.     

Vascularization of the myocutaneous latissimus dorsi flap. Injection study on the thoracodorsal artery

This study reports on investigations of the thoracodorsal artery by injection studies. This artery has a long proximal extramuscular course before it enters the muscle. A consistent neurovascular hilum was found at a considerably constant location on the inferior muscle surface, 2-3 cm medial to its lateral edge and about 5 cm distal to the inferior scapular border. A constant branching pattern of the thoracodorsal artery into a lateral and a medial vessel stem was found. A high number of muscle-perforating arteries from a dense network in the cutaneous and subcutaneous layer. The thoracodorsal artery supplies the whole cutaneous area adjacent to the latissimus dorsi muscle plus a streak of about 2 cm at the medial and distal muscle borders. The presented anatomical landmarks are useful for locating the neurovascular hilum, and the intramuscular course of the thoracodorsal artery for exploration of the vessel by Doppler sonography or dissection. The length of the pedicle and the relatively big vessel gauge are good anatomical markers for the free transfer of a latissimus dorsi flap.     

An anomalous muscle (accessory subscapularis-teres-latissimus muscle) in the axilla penetrating the brachial plexus in man.

An anomalous muscle passing through the brachial plexus was found in 10 cases out of 380 sides of 190 human cadavers in the dissection course. The muscle was designated as 'accessory subscapularis-teres-latissimus muscle'. This muscle arose near the lateral margin of the scapula, either from the surface of the subscapularis muscle or from the border of the quadrangular terminal tendon of the latissimus dorsi or from both of those sources when the muscle was divided into two heads. It ran obliquely upward to fuse with the insertion of the subscapularis. The largest anomaly was 2.5 cm in width and 7 cm in length. This muscle could be classified into three types on the basis of its nerve supply and its relation to the brachial plexus. The type I muscle crossed over the axillary and lower subscapular nerves, behind the radial nerve and was innervated by the lower subscapular nerves. The type II musclepenetrated the brachial plexus separating the radial nerve into two roots; the upper from the posterior division of the upper trunk and the lower from the posterior divisions of the middle and lower trunks. The type II muscle was supplied by a branch of the radial nerve, which originated always at the same level as the origin of the thoracodorsal nerve. The type III muscle passed through the further more ventrocaudal level of the plexus; in one case it divided the radial nerve into an upper root from the posterior divisions of the upper and middle trunks and a lower root from the lower trunk, and, in another case, into an upper main root from all the three trunks and a lower slender root from the lower trunk. The type III muscle was supplied by branches from the radial and in addition from the thoracodorsal nerve in one case. In four out of ten cases, the subscapular or thoracodorsal artery also passed posterior to the anomalous muscle. A discussion was made on the nature of the anomalous muscle.     

Comparative evaluation of scapular and humeral coordinate systems based on biomedical images of the glenohumeral joint

Bio-imaging techniques represent a powerful tool for shoulder joint biomechanical analysis. However, the restricted field of view may prevent the acquisition of complete scapula and humerus bone models and hence limiting the applicability of standardized anatomical coordinate system (ACS) definitions. The aim of this study was to propose ACS definitions for both scapula and humerus which can be implemented when limited portions of the relevant bones are available. Magnetic resonance (MR) images of twenty right humeri and scapulae were acquired. The proposed ACSs were assessed in terms of (1) sensitivity to bone morphological variation, (2) intra – and inter – operator repeatability and (3) consistency with the anatomical cardinal directions. A comparison with alternative ACS definitions was also performed. Overall, our ACS scapular proposal and that presented in Kedgley and Dunning (2010) were found to be the least sensitive to the morphometric variability (mean angular absolute deviation lower than 8.3 deg) and they were characterized by a high intra – and inter – operator repeatability (mean angular absolute deviation lower than 1.5 deg). The humeral ACS proposal showed a morphometric variability similar to Amadi et al. (2009b) (mean angular absolute deviation lower than 8.3 deg) but a higher reproducibility. The scapular and humeral ACS mean angular deviation from the reference anatomical cardinal directions were smaller than 15 deg and 8.6 deg, respectively. The proposed scapular and humeral ACS definitions are therefore suitable to be applied when a limited portion of the glenohumeral joint is available as it may occur in standard shoulder clinical exams.     

Bone remodelling of the scapula after a total shoulder arthroplasty

According to Wolff’s law, the changes in stress after a prosthesis implantation may modify the shape and internal structure of bone, thus compromising the long-term prosthesis fixation and, consequently, be a significant factor for glenoid loosening. The aim of the present study is to evaluate the changes in the bone adaptation process of the scapula after an anatomical and reverse total shoulder arthroplasty. Five finite element models of the implanted scapula are developed considering the implantation of three anatomical, cemented, all-polyethylene components; an anatomical, cementless, metal-backed component; and a reverse, all-metal component. The methodology followed to simulate the bone adaptation of the scapula was previously validated for the intact model, prior to the prosthesis implantation. Additionally, the influence of the bone quality on the adaptation process is also investigated by considering an osteoporotic condition. The results show that the stress shielding phenomenon is more concerning in cementless, metal-based components than in cemented, all-polyethylene components, regardless of the bone quality. Consequently, as far as the bone adaptation process of the bone is concerned, cemented, all-polyethylene components are better suited for the treatment of the shoulder joint.     

Applied anatomy of the anterolateral femoral flap

A study of the source of the blood supply to the anterolateral femoral flap was carried out on 42 lower limbs of adult cadavers (among them 35 cadavers with injection of red latex and 1 with india ink into the arteries and 6 vascular cast specimens), and the surface locations of the vascular pedicle were detected on 50 healthy adults. It was found that the descending branch of the lateral circumflex femoral vessel was an ideal axial vessel. There are constant perforating branches of the myocutaneous artery or cutaneous branches from the intermuscular space to the anterolateral femoral skin. The area extends about 12 x 30 cm. Within the flap, the anterior branch of the anterolateral cutaneous nerve of the high is located. This flap has been widely used for free transplantation in China since 1983 with satisfactory results.     

Outcomes of scapula stabilization in obstetrical brachial plexus palsy: a novel dynamic procedure for correction of the winged scapula

Among the late consequences of obstetrical brachial plexus palsy is winging of the scapula, a functional and aesthetic deformity. This article introduces a novel surgical procedure for the dynamic correction of this clinical entity that involves the dynamic transfer of the contralateral trapezius muscle and/or rhomboid muscles and anchoring to the affected scapula. In more severe cases of scapula winging, the contralateral latissimus dorsi muscle may also need to be transferred to achieve dynamic scapula stabilization. The outcomes of this novel surgical procedure were analyzed in relation to the effect on abduction, external rotation, growth of the scapula, and distance of the scapula from the posterior midline. The results were analyzed in 26 patients who underwent this procedure and had adequate follow-up. The mean patient age was 6.39 years. Fourteen (54 percent) had a diagnosis of Erb palsy, and 12 (46 percent) had a diagnosis of global paralysis. All 26 patients had an additional secondary procedure performed prior to or simultaneously with the scapula stabilization procedure. In 19 patients, the contralateral trapezius was transferred and anchored to the medial border of the winged scapula alone, but in seven cases the underlying rhomboid major was transferred along with the trapezius muscle to provide sufficient scapula stabilization. In five cases in which the scapula winging was severe, the contralateral latissimus dorsi muscle was transferred at a second stage. After this procedure, all patients demonstrated improved scapula symmetry. The mean increase in abduction was 18 degrees (p < 0.001), the mean increase in external rotation was 19 degrees (p < 0.001), and the mean increase in anterior flexion was 12 degrees (p = 0.015). The improvement of the relative position of the winged scapula on the posterior thorax was analyzed by measuring the distance of the inferior angle of both scapulae from the midline, then calculating the difference between normal and affected sides and comparing this value before and after the scapula stabilization procedure. This value preoperatively was 3.24 cm; postoperatively it decreased to 0.36 cm (p < 0.001), demonstrating a statistically significant improvement.     

The thoracodorsal artery perforator flap: clinical experience and anatomic study with emphasis on harvest techniques

The thoracodorsal artery perforator flap is a relatively new flap that has yet to find its niche in reconstructive surgery. At the authors' institution it has been used for limb salvage, head and neck reconstruction, and trunk reconstruction in cases related to trauma, burns, and malignancy. The authors have found the flap to be advantageous for cranial base reconstruction and for resurfacing the face and oral cavity. The flap has been used successfully for reconstruction of traumatic upper and lower extremity defects, and it can be used as a pedicled flap or as a free tissue transfer. The perforating branches of the thoracodorsal artery offer a robust blood supply to a skin-soft-tissue paddle of 10 to 12 cm x 25 cm, overlying the latissimus dorsi muscle. The average pedicle length is 20 cm (range, 16 to 23 cm), which allows for a safe anastomosis outside the zone of injury in traumatized extremities; the flap can be made sensate by neurorrhaphy with sensory branches of the intercostal nerves. Vascularized bone can be transferred with this flap by taking advantage of the inherent vascular anatomy of the subscapular artery. A total of 30 pedicled and free flap transfers were performed at the authors' institution with an overall complication rate of 23 percent and an overall flap survival rate of 97 percent. Major complications, such as vascular thrombosis, return to the operating room, fistula formation, recurrence of tumor, and flap loss, occurred in 17 percent of the patients. Despite these drawbacks, the authors have found the thoracodorsal artery perforator flap to be a safe and extremely versatile flap that offers significant advantages in acute and delayed reconstruction cases.     

The supply of blood in the skin territory above the lower part of the serratus anterior muscle

At present, the putative clinical use of the musculocutaneous and ostomusculocutaneous serratus anterior flaps has been compromised by the risk of partial or total necrosis of the skin overlying the lower part of the serratus anterior muscle. Therefore, the aim of this study was to delineate a skin area vascularized by perforant musculocutaneous branches of arteries stemming from the lower segment of the anterior serrated muscle. Black ink was injected in thoracodorsal artery branches for the serratus anterior muscle in 50 human cadavers before the autopsies (the study was approved by the Institutional Review Board). The surface area of the labeled skin was determined and its borders delineated by means of transparent millimeter grid. Planimetry data were subsequently analyzed with the aid of PC computer program. The results show that the calculated mean surface area (143.79 +/- 2.68 x 2.077; range 138.22-149.36 cm2) of the skin vascularized by perforant musculocuaneous branches stemming from the lower segment of the anterior serrated muscle, can serve as a reliable guide for taking serratus anterior flap in any patient. Therefore, appropriately sized musculocutaneous or osteomusculocutaneous serratus anterior flap can be safely and efficiently used in plastic and reconstructive surgery.     

The osteocutaneous scapular flap for mandibular and maxillary reconstruction

Microfil injections in 8 cadavers and clinical experience with 26 patients have demonstrated a reliable blood supply to the lateral border of the scapula based on branches of the circumflex scapular artery. This tissue has been used successfully for reconstruction of a variety of defects resulting from maxillectomy and mandibular defects from cancer and benign tumor excisions. Advantages of this tissue over previous reconstructive methods include the ability to design multiple cutaneous panels on a separate vascular pedicle from the bone flap allowing improvement in three-dimensional spatial relationships for complex mandibular and maxillary reconstructions. The lateral border of the scapula provides up to 14 cm of thick, straight corticocancellous bone that can be osteotomized where desired. The thin blade of the scapula provides optimum tissues for palate and orbital floor reconstruction. There have been no flap failures and minimal donor-site complications.     

Subject-specific bone remodelling of the scapula

Finite element analyses, with increasing levels of detail and complexity, are becoming effective tools to evaluate the performance of joint replacement prostheses and to predict the behaviour of bone. As a first step towards the study of the complications of shoulder arthroplasty, the aim of this work was the development and validation of a 3D finite element model of an intact scapula for the prediction of the bone remodelling process based on a previously published model that attempts to follow Wolff's law. The boundary conditions applied include full muscle and joint loads taken from a multibody system of the upper limb based on the same subject whose scapula was here analysed. To validate the bone remodelling simulations, qualitative and quantitative comparisons between the predicted and the specimen's bone density distribution were performed. The results showed that the bone remodelling model was able to successfully reproduce the actual bone density distribution of the analysed scapula.     

Redescription of Rayososaurus agrioensis Bonaparte (Sauropoda, Diplodocoidea), a rebbachisaurid from the early Late Cretaceous of Neuquén

The material belonging to the holotype of Rayososaurus agrioensis Bonaparte is redescribed. The great development of the acromion process, directed in a markedly posterior direction, and the ventral margin of the scapula with a strong ventrodistal expansion, interpreted as autapomorphies of the taxon, justify the validity of the species. Although the material is rather fragmentary, the racquet-shaped scapula, the distal expansion positioned at the same height as the proximal expansion, and the well-developed acromion process allow the taxon to be included within Rebbachisauridae. At the same time, the angle of less than 90? between the coracoid articulation and the scapular blade, together with the great development of the acromion process of the scapula, justify its inclusion in a clade situated as a sister group of Cathartesaura. On the basis of the stratigraphic calibration of the phylogenetic analysis, an important diversification event of the rebbachisaurids is deduced during the Hauterivian-Barremian. In this context, the presence of two monophyletic groups, one of which comprises African-European taxa and the other South American taxa, would place the definitive separation of South America and Africa within this time interval.     

The thoracoacromial vessels as recipient vessels in microsurgery and supermicrosurgery: an anatomical and sonographic study

The purpose of this study was to investigate the presence and the precise course of the pectoral branch of the thoracoacromial vessels on the underside of the pectoralis major muscle by anatomical dissection and by color Doppler ultrasound. A further goal was to determine whether these vessels were suitable as recipient vessels in microsurgery and supermicrosurgery for breast reconstruction. In 18 cadavers, the pectoral branch of the thoracoacromial vessels was followed caudally until the diameter of the artery diminished to 1 mm. The same examination was carried out in 40 young female volunteers by tracing the vessel course with color Doppler ultrasound. The 1-mm cutoff point of the artery was measured with reference to the manubrium, the midsternal line, the clavicle, and the upper border of the closest rib. In addition, in the cadavers, the 2-mm cutoff point was determined. At both cutoff points, the diameter of the accompanying vein was measured. The pectoral branch of the thoracoacromial vessels with the artery and concomitant veins could be detected on all 100 undersides of the pectoralis major muscle, anatomically and sonographically. In their course from the acromial region downward, the arteries reached a diameter of 1 mm at an average of 9.9 cm from the manubrium, horizontally 9.4 cm from the midsternal line, and vertically 4.0 cm from the lower border of the clavicle. The 1-mm reference point was situated on the upper border of the third rib in 85 percent of cases. The average distance between the 1-mm and the 2-mm cutoff points was 3.5 cm. At the 1-mm cutoff point, the diameter of the vein was 0.9 mm, and at the 2-mm cutoff point, it was 1.7 mm. Because of their central position at the anterior hemithorax, these vessels are easily accessible from mastectomy incisions, even in skin-sparing mastectomies; the donor-site morbidity is negligible; and as the diameters of the vessels gradually decrease along their caudal course, the recipient site can be chosen precisely according to the length and the diameter of the donor vessels and major mismatch can be avoided. Thus, the pectoral branches of the thoracoacromial vessels are well suited as recipient vessels for (super)microsurgery and are a very promising addendum to the thoracodorsal and internal mammary vessels.     

Direct kinematic modeling of the upper limb during trunk-assisted reaching

The study proposes a rigid-body biomechanical model of the trunk and whole upper limb including scapula and the test of this model with a kinematic method using a six-dimensional (6-D) electromagnetic motion capture (mocap) device. Large unconstrained natural trunk-assisted reaching movements were recorded in 7 healthy subjects. The 3-D positions of anatomical landmarks were measured and then compared to their estimation given by the biomechanical chain fed with joint angles (the direct kinematics). Thus, the prediction errors was attributed to the different joints and to the different simplifications introduced in the model. Large (approx. 4 cm) end-point prediction errors at the level of the hand were reduced (to approx. 2 cm) if translations of the scapula were taken into account. As a whole, the 6-D mocap seems to give accurate results, except for prono-supination. The direct kinematic model could be used as a virtual mannequin for other applications, such as computer animation or clinical and ergonomical evaluations.     

Accuracy of X-Ray and Magnetic Resonance Imaging in Defining the Tumor Margin in Primary Bone Sarcoma

BackgroundLimb-salvage surgery for primary bone sarcomas are preceded by X-ray and MRI for surgical planning. However, the accuracy of X-ray and MRI predicted margins are not well described. Our study examined these questions: (1) How accurately do X-ray and MRI margin measurements reflect the true margin on pathology reports? (2) Do X-ray or MRI margin measurements have smaller differences compared to pathology reports? (3) How many X-ray or MRI margin measurement differences were greater than 1 cm, 2 cm, and 3 cm from pathology reports? (4) Is there an X-ray or MRI view that consistently results in a smaller difference from pathology reports?MethodsThis retrospective chart review examined patients with primary bone sarcoma treated with limb-salvage surgery. Reviewers used electronic measurement tools to determine margins from X-ray or MRI based on the resection length of the pathologic specimen. Mean differences of margin measurements to pathology reports were calculated. We determined outliers of imaging margin measurements at 1 cm, 2 cm, and 3 cm differences to pathology reports.ResultsIn the total cohort of 39 patients, the mean difference of X-ray and MRI margins compared to pathology reports were 1.09 cm (st dev 0.79 cm) and 0.71 cm (st dev 0.70 cm), respectively. MRI margin measurements had smaller differences compared to pathology reports than X-ray in 32 of 38 cases (84%) with complete imaging. X-ray outliers at 1 cm, 2 cm, and 3 cm differences were 36, 14 and 2 respectively for 70 margin measurements and MRI outliers at 1 cm, 2 cm, and 3 cm differences were 17, 6, and 0 respectively for 66 margin measurements. The views with the smallest difference were anterior-posterior X-rays and MRI views with the closest predicted margin.ConclusionElectronic MRI margin measurements with the closest predicted margin provided the smallest differences with pathology reports and are therefore the most accurate for preoperative planning. When there is adequate residual diaphysis for reconstructive fixation, surgeons should plan for a 3 cm bone margin using MRI measurements to ensure complete removal of the intramedullary extent of sarcoma.Level of Evidence: IV     

Myocutaneous vascular territory of the thoracoacromial artery. A topographical and morphometric study of the arterial vascularization of the pectoralis major myocutaneous flap

In order to clarify the vascularization of the pectoralis major muscle by the pectoral branch of the thoracoacromial artery, morphometric investigations and selective injection studies of the vessel were performed. A comparison with the anatomical orientation points clinically used for locating the vessel and its course confirms that the origin of the artery is at a constant spot at half the sternoacromial distance; the main vessel axis, however, deviates considerably from the clinically used acromioxiphoid axis in lateral direction. The lumen of the thoracoacromial artery has a range of more than 150% of the smallest vessel. Atherosclerotic alterations are not the reason for this variety but only an additional handicap for the vessel capacity. Two types of muscle-perforating arteries were found, and the cutaneous area of blood supply was defined.     

A clinical experience with perforator flaps in the coverage of extensive defects of the upper extremity

Traditional skin free flaps, such as radial arm, lateral arm, and scapular flaps, are rarely sufficient to cover large skin defects of the upper extremity because of the limitation of primary closure at the donor site. Muscle or musculocutaneous flaps have been used more for these defects. However, they preclude a sacrifice of a large amount of muscle tissue with the subsequent donor-site morbidity. Perforator or combined flaps are better alternatives to cover large defects. The use of a muscle as part of a combined flap is limited to very specific indications, and the amount of muscle required is restricted to the minimum to decrease the donor-site morbidity. The authors present a series of 12 patients with extensive defects of the upper extremity who were treated between December of 1999 and March of 2002. The mean defect was 21 x 11 cm in size. Perforator flaps (five thoracodorsal artery perforator flaps and four deep inferior epigastric perforator flaps) were used in seven patients. Combined flaps, which were a combination of two different types of tissue based on a single pedicle, were needed in five patients (scapular skin flap with a thoracodorsal artery perforator flap in one patient and a thoracodorsal artery perforator flap with a split latissimus dorsi muscle in four patients). In one case, immediate surgical defatting of a deep inferior epigastric perforator flap on a wrist was performed to immediately achieve thin coverage. The average operative time was 5 hours 20 minutes (range, 3 to 7 hours). All but one flap, in which the cutaneous part of a combined flap necrosed because of a postoperative hematoma, survived completely. Adequate coverage and complete wound healing were obtained in all cases. Perforator flaps can be used successfully to cover a large defect in an extremity with minimal donor-site morbidity. Combined flaps provide a large amount of tissue, a wide range of mobility, and easy shaping, modeling, and defatting.     

Difference between palpation and optoelectronics recording of scapular motion

The aim of this study is to determine the errors of scapular localisation due to skin relative to bone motion with an optoelectronic tracking system. We compared three-dimensional (3D) scapular positions obtained with skin markers to those obtained through palpation of three scapular anatomical landmarks. The scapular kinematics of nine subjects were collected. Static positions of the scapula were recorded with the right arm elevated at 0°, 40°, 80°, 120° and 160° in the sagittal plane. Palpation and subsequent digitisation of anatomical landmarks on scapula and thorax were done at the same positions. Scapular 3D orientation was also computed during 10 repeated movements of arm elevation between 0° and 180°. Significant differences in scapular kinematics were seen between static positions and palpation when considering anterior/posterior tilt and upward/downward rotation at angles over 120° of humeral elevation and only at 120° for internal/external rotation. There was no significant difference between positions computed during static positions and during the movement for the three scapular orientations. A rotation correction model is presented in order to reduce the errors between static position and palpation measurement.     

Properties of the celiac trunk--anatomical study

Although anatomical properties and vessel variations of the celiac trunk are well explored in the literature, there is not so much information on the arterial diameters, and this data is important for surgical procedures and angiographic examinations. The aim of this study was to investigate properties of the celiac trunk in humans by using anatomical dissection. Ninety cadavers were dissected for the celiac trunk identification and arterial diameter measurements. The results of anatomical examination showed that in 72% of all cases the celiac trunk divides into the splenic artery and the common hepatic artery, while the left gastric artery arises as a first branch and had origin between aorta, all over the celiac trunk up to a bifurcation. From the 90 cadavers, 4 presented anatomical variations. Where normal anatomy was present, the mean length of the celiac trunk was 1.9 +/- 0.08 cm and its mean arterial diameter was 0.78 +/- 0.08 cm. The splenic artery had the largest diameter (0.61 +/- 0.05 cm) and the left gastric artery had the smallest diameter (0.38 +/- 0.03 cm). Our data represent original results about anatomical variations and arterial diameter of the celiac trunk and its main branches provided by anatomical dissection.     

Fascia canals of the greater sciatic foramen and their practical importance

Fasciae and fat tissue spaces in the gluteal region, topography of the suprapiriform and infrapiriform foramina have been studied by means of a complex anatomical experimental technique. The suprapiriform foramen should be considered as a fascialosseous canal, as it is 4-4.5 cm long and 0.6-1.0 cm wide. It is formed by the upper margin of the greater sciatic notch covered with a thin fascia, fasciae of the gluteal and piriform muscles and the parietal layer of the pelvic fascia. The proper fascial vaginae of the upper gluteal vessels and nerves are adhered to fascial walls of the canal. This peculiarity is used for the method of ligation of the superior gluteal artery within the limits of the suprapiriform canal. The infrapiriform foramen is either narrow or wide enough (up to 2.0 cm in diameter). Inferior gluteal vessels at the level of the sacrospinous ligament go from the parietal layer of the pelvic fascia into the duplicature of the deeper layer of musculus gluteus maximus. The inferior gluteal nerve, above the lower margin of the piriform muscle, ajoining the vessels gets into the fissure of the parietal layer of the pelvic fascia, under the lower margin neurovascular fasciculus also goes through the fissure of the pelvic fascial parietal layer, downward and parallel to the inferior gluteal vessels. The knowledge of possible ways of connections through the canals of the greater sciatic foramen, fat tissue spaces at the subperitoneal level of the small pelvis and the gluteal region is of great practical value.